SI Joint Radiofrequency Ablation

At Modal Pain Management NYC in Midtown Manhattan, Dr. Alex Movshis performs image-guided sacroiliac (SI) joint radiofrequency ablation — lateral branch RFA of the L5 dorsal ramus and the S1–S3 lateral branches — as the durable treatment step for confirmed SI joint pain. The sacroiliac joint accounts for an estimated 15–30% of chronic low back pain centered below L5, according to the 2025 multisociety guideline, a range set by early diagnostic block studies (Schwarzer 1995, Maigne 1996), and for many of those patients the treatment story follows the same arc: a diagnostic SI joint injection finally identifies the pain source after months of unrevealing lumbar MRIs, the injection works — and then it wears off within weeks to a few months, again and again. Lateral branch radiofrequency ablation breaks that cycle. By interrupting the small sensory nerves that carry pain from the back of the joint rather than repeatedly medicating the joint itself, a single 45–60 minute procedure gives relief that commonly lasts about 6 months to a year in people who respond, and it can be repeated when pain returns. Located at 369 Lexington Avenue Floor 25 in NYC 10017, we perform the procedure under live fluoroscopy using lesion techniques matched to the published evidence — and at a flat, all-inclusive self-pay price of $1,500 for patients paying directly.

Why the SI Joint Needs Its Own Ablation Technique

The sacroiliac joint is not innervated like the lumbar facet joints, and that anatomic difference is the single most important fact about this procedure. A lumbar facet joint receives its pain supply from two medial branch nerves that sit in predictable bony grooves — which is why facet radiofrequency ablation after a positive medial branch block is so reproducible. The posterior SI joint, by contrast, is supplied by the L5 dorsal ramus and the lateral branches of the S1, S2, and S3 dorsal sacral rami — a fan of small nerves that exit the sacral foramina and vary in number, depth, and position from person to person, and even from side to side in the same person.

That variability killed the naive version of this procedure. A single conventional lesion placed where the lateral branches “should” be will miss them in a meaningful fraction of patients. In a sham-controlled trial of 60 patients that used conventional heat lesions, ablation did no better than sham at 3 months (van Tilburg 2016). The techniques that produced positive randomized results solve the coverage problem in one of two ways. Cooled-tip radiofrequency circulates cool water through the electrode, which produces a large spherical lesion at each target, and it is the technique used in most of the trials below. Bipolar “palisade” strip lesioning places a row of closely spaced cannulae along the dorsal sacrum lateral to the foramina and creates a continuous strip lesion that the variable branches cannot slip through. Modal Pain uses bipolar palisade strip lesioning. It belongs to the large-lesion group that the 2025 multisociety guideline rates as having the stronger evidence, and its own trials are summarized below.

The Diagnostic Gateway: No Ablation Without a Positive Block

SI joint RFA is only as good as the diagnosis in front of it, and Modal Pain does not perform the procedure without block confirmation. Two pathways lead in. The first is the image-guided intra-articular SI joint injection — if the local anesthetic phase relieves enough of the typical buttock and low-back pain during the anesthetic window to meet the threshold your plan uses, the joint becomes a likely pain generator. A positive joint injection does not reliably predict how well ablation will work, though, because the ablation reaches the nerves to the back of the joint and its ligaments, not the whole joint. The 2025 multisociety guideline uses 50% relief, while some societies and insurers require 75% or more on two blocks. The second, used when RFA is the likely next step or when the insurer’s policy requires it, is a set of diagnostic lateral branch blocks: small volumes of local anesthetic placed directly on the L5 dorsal ramus and S1–S3 lateral branches — the exact nerves the ablation would treat. Because the block numbs the same nerves the ablation treats, the 2025 multisociety guideline prefers lateral branch blocks for selecting ablation candidates, though how well any block predicts the result is still uncertain.

This gateway is not bureaucratic caution. In every positive trial of the procedure, patients were selected by block response, and block-selected patients are the population in which the success rates below were observed. Skipping the diagnostic step means operating on a guess.

The Evidence, Honestly Read

The randomized evidence for lateral branch RFA is mixed. Two small sham-controlled trials were clearly positive, a large comparative trial was positive against standard care, and a large pragmatic trial and a recent meta-analysis found smaller effects. It also comes with a lesson about technique that an honest clinic should explain rather than hide.

Cohen 2008 (Anesthesiology). The first placebo-controlled randomized trial: 28 patients with injection-confirmed SI joint pain received either lateral branch denervation (conventional lesions of the L4 and L5 dorsal rami, cooled-probe lesions of the S1–S3 lateral branches) or a sham procedure. At 1, 3, and 6 months, 79%, 64%, and 57% of ablated patients had 50%+ pain relief with significant functional improvement, versus 14% of placebo patients at 1 month and none at 3 months. By one year, 2 of the 14 treated patients still had relief.

Patel 2012 (Pain Medicine) and the 12-month follow-up (Pain Practice). Fifty-one patients with dual-positive lateral branch blocks were randomized 2:1 to cooled lateral branch neurotomy or sham. At 3 months, 47% of treated patients versus 12% of sham patients met the composite treatment-success criteria, with statistically significant gains in pain, disability, physical function, and quality of life. The treated group’s success rate was 38% at 6 months and 59% at 9 months. At 12 months, the follow-up publication reported a mean 2.7-point drop in pain scores, a 13.9-point improvement in Oswestry disability score, and better SF-36 bodily pain scores compared with before treatment. Those are changes from baseline, not comparisons with sham, because sham patients were allowed to cross over to treatment after 3 months.

The negative trial. The 2017 Dutch MINT trials (Juch, JAMA) added radiofrequency denervation to an exercise program in 228 patients with SI joint pain selected by at least 50% relief from diagnostic lateral branch blocks. At 3 months, pain in the ablation group was 0.71 points lower on a 0–10 scale, a statistically significant difference that fell short of the trial’s 2-point threshold for a clinically important benefit. The trial was pragmatic and unblinded, with no sham procedure, and its result sits against the sham-controlled trials above. The 2025 multisociety guideline weighed all of these trials and concluded that ablation with anatomically valid techniques (cooled-tip or bipolar strip lesions) improves pain and disability for at least 6 months, with stronger evidence for larger lesions. We follow the same approach: block-based selection and a lesion strategy that covers the lateral branch territory.

The largest trial. Cohen 2024 (Regional Anesthesia & Pain Medicine) randomized 210 patients, selected by diagnostic SI joint injections and lateral branch blocks, to cooled RFA or standard medical management (medications, injections, physical therapy, and integrative care). There was no sham procedure, and the trial was sponsored by Avanos, the maker of the cooled-RF system. At 3 months, pain had fallen by an average of 2.5 points with cooled RFA versus 0.4 points with standard care, and 52.3% versus 4.3% were responders (at least a 2-point or 30% drop in pain plus feeling at least moderately better). At 12 months, 57.4% (35 of 61) of the cooled-RFA group still had at least a 2-point or 30% drop in pain from baseline (Cohen 2026). Because the comparison was standard care rather than a sham, part of that difference can reflect expectation.

The pooled picture. A 2026 meta-analysis of randomized trials in BMJ Open (Truong 2026) found that RF denervation for SI joint pain lowered pain more than sham by about 1.3 points at one month, with a confidence interval that included no effect, and by about 1.5 points over 1 to 6 months, based on two trials with 68 patients. The certainty was low, and both effects were smaller than the review’s threshold for a clinically important difference. We share these numbers because they are part of an honest decision: ablation helps a good share of carefully selected patients, and it does not help everyone.

Bipolar strip-lesion data. In a double-blind, sham-controlled trial of patients whose SI joint pain was confirmed by diagnostic blocks, average pain in the 11 patients treated with a bipolar strip-lesion device fell from 8.1 to 3.4 out of 10 at 3 months, while the 6 sham patients went from 7.3 to 7.0, and results were similar at 6 months (Mehta 2018). The trial was small and was stopped early after an interim analysis showed the difference. A randomized comparison by Cánovas Martínez and colleagues (2016) found that bipolar palisade RFA of the S1–S3 lateral branches outperformed intra-articular SI joint injections in 60 patients. The injection group had no significant pain reduction at 3 or 12 months, and the wider-spacing palisade group still had lower pain scores at 12 months than the injection group, consistent with the strip-lesion logic above.

A guideline that recommends against it. A 2025 clinical practice guideline in The BMJ (Busse 2025) made a strong recommendation against radiofrequency ablation of the facet or sacroiliac joints for chronic axial spine pain lasting 3 months or more, based on a linked systematic review and network meta-analysis of the randomized trials. It does not apply to pain under 3 months or to inflammatory arthritis. The 2025 multisociety guideline, from a working group drawn from 25 organizations, read the same trials as showing relief for at least 6 months with large-lesion techniques in patients selected by blocks (McCormick 2025). The two guidelines weigh the small sham-controlled trials and the larger pragmatic trials differently. We share both so that the decision is made with the full picture.

What the Procedure Is Like

Before. You arrive without needing to fast (unless IV sedation is planned), medications reviewed in advance — your blood thinner plan is made with you in advance, following ASRA periprocedural guidelines. The skin over the sacrum is prepped and numbed with lidocaine.

During. Under live fluoroscopy, Dr. Movshis positions the radiofrequency cannulae at the L5 dorsal ramus and along the lateral border of the S1–S3 posterior sacral foramina. Stimulation testing checks that the electrode sits safely away from nerves that control movement — the lateral branches carry no motor fibers to the legs. Each lesion cycle runs about 90 seconds. You’ll feel deep pressure or a brief ache over the sacrum during the cycles. Patients who had sedation at the surgical center typically remember little of it. Total room time is 45–60 minutes.

After. A short observation period, then home — with a driver if you had sedation, on your own if not. Soreness over the treated area lasts 3–7 days and responds to ice and acetaminophen. A temporary flare of pain in the first one to two weeks is common, sometimes from neuritis, and it usually settles without treatment. Normal activity resumes in 24–48 hours, and the real benefit declares itself over 2–4 weeks. We formally measure the response at a 4–6 week follow-up and then track durability at 3, 6, and 12 months so the repeat cycle — if one is ever needed — is planned rather than improvised.

The full aftercare instructions are on the radiofrequency ablation aftercare page.

Cost: $1,500 Flat, or Insurance When Your Plan Covers It

For patients paying directly, SI joint radiofrequency ablation at Modal Pain Management NYC is a flat, all-inclusive $1,500 per session — physician fee, fluoroscopic guidance, and medication included, no separate facility fee. It sits alongside every other procedure price on our self-pay pricing page, and 0%-interest monthly payment plans through Cherry can spread the cost. The arithmetic worth doing: a $700 SI joint injection whose relief fades within two months costs more per month of relief than a $1,500 ablation that lasts six months or longer, which is why patients who respond to ablation often come out ahead of a repeat-injection cycle. Not everyone responds, which is why the diagnostic blocks come first.

For insured patients, coverage for lateral branch RFA varies by plan. Several major insurers currently classify it as unproven or investigational, and coverage is decided case by case, so Modal Pain checks your plan’s specific policy before scheduling and obtains prior authorization when the plan covers it. Verify your benefits before deciding which route costs you less.

Who Should — and Shouldn’t — Have This Procedure

The strongest candidates share three features. Their SI joint pain has been traced by a positive diagnostic block. They get a good but short-lived response from SI joint injections. And the pain is chronic and mechanical, including the common post-lumbar-fusion pattern, where the SI joint absorbs extra load after the spine above it is fixed (Cohen 2013), and where the SI joint is estimated to be the pain source in 33% to 59% of people with low back pain after lumbar or lumbosacral fusion (according to the 2025 multisociety guideline).

The procedure is the wrong tool for inflammatory sacroiliitis from ankylosing spondylitis or psoriatic arthritis, where the driver is systemic and the foundation of care is rheumatologic disease-modifying therapy, not denervation. It is also not first-line for acute SI joint pain — many recent-onset cases settle with physical therapy and, when needed, a single injection. And patients with true structural instability on imaging may need a surgical fusion conversation instead, and Modal Pain coordinates that referral when the ladder points there.

If your low-back or buttock pain has already been traced to the SI joint — or you suspect it should be — the consultation visit sorts out where you are on that ladder and whether ablation is your next step. Book online or call (646) 290-6660.

References

• Cohen SP, Hurley RW, Buckenmaier CC 3rd, Kurihara C, Morlando B, Dragovich A. Randomized placebo-controlled study evaluating lateral branch radiofrequency denervation for sacroiliac joint pain. Anesthesiology. 2008;109(2):279–88. doi:10.1097/ALN.0b013e31817f4c7c · PubMed
• Patel N, Gross A, Brown L, Gekht G. A randomized, placebo-controlled study to assess the efficacy of lateral branch neurotomy for chronic sacroiliac joint pain. Pain Med. 2012;13(3):383–98. doi:10.1111/j.1526-4637.2012.01328.x · PubMed
• Patel N. Twelve-Month Follow-Up of a Randomized Trial Assessing Cooled Radiofrequency Denervation as a Treatment for Sacroiliac Region Pain. Pain Pract. 2016;16(2):154–67. doi:10.1111/papr.12269 · PubMed
• Juch JNS, Maas ET, Ostelo RWJG, et al. Effect of Radiofrequency Denervation on Pain Intensity Among Patients With Chronic Low Back Pain: The Mint Randomized Clinical Trials. JAMA. 2017;318(1):68–81. doi:10.1001/jama.2017.7918 · PubMed
• Cánovas Martínez L, Orduña Valls J, Paramés Mosquera E, Lamelas Rodríguez L, Rojas Gil S, Domínguez García M. Sacroiliac joint pain: Prospective, randomised, experimental and comparative study of thermal radiofrequency with sacroiliac joint block. Rev Esp Anestesiol Reanim. 2016;63(5):267–72. doi:10.1016/j.redar.2015.08.003 · PubMed
• Cohen SP, Kapural L, Kohan L, et al. Cooled radiofrequency ablation versus standard medical management for chronic sacroiliac joint pain: a multicenter, randomized comparative effectiveness study. Reg Anesth Pain Med. 2024;49(3):184–191. doi:10.1136/rapm-2023-104568 · PubMed
• Cohen SP, Kapural L, Kohan L, et al. Cooled radiofrequency ablation provides extended clinical utility in the management of chronic sacroiliac joint pain: 12-month follow-up results from the observational phase of a randomized, multicenter, comparative-effectiveness crossover study. Reg Anesth Pain Med. 2026;51(6):660–669. doi:10.1136/rapm-2024-106315 · PubMed
• Truong KMTL, Maas E, Rasmussen MM, Ostelo R, Hurri H, van Tulder MW. Current evidence on radiofrequency denervation for chronic low back pain: a systematic review and meta-analysis. BMJ Open. 2026;16(7):e105106. doi:10.1136/bmjopen-2025-105106 · PubMed
• Mehta V, Poply K, Husband M, Anwar S, Langford R. The Effects of Radiofrequency Neurotomy Using a Strip-Lesioning Device on Patients with Sacroiliac Joint Pain: Results from a Single-Center, Randomized, Sham-Controlled Trial. Pain Physician. 2018;21(6):607–618. PubMed
• van Tilburg CWJ, Schuurmans FA, Stronks DL, Groeneweg JG, Huygen FJPM. Randomized Sham-controlled Double-Blind Multicenter Clinical Trial to Ascertain the Effect of Percutaneous Radiofrequency Treatment for Sacroiliac Joint Pain: Three-month Results. Clin J Pain. 2016;32(11):921–926. doi:10.1097/AJP.0000000000000351 · PubMed
• Busse JW, Genevay S, Agarwal A, et al. Commonly used interventional procedures for non-cancer chronic spine pain: a clinical practice guideline. BMJ. 2025;388:e079970. doi:10.1136/bmj-2024-079970 · PubMed
• Cohen SP, Chen Y, Neufeld NJ. Sacroiliac joint pain: a comprehensive review of epidemiology, diagnosis and treatment. Expert Rev Neurother. 2013;13(1):99–116. doi:10.1586/ern.12.148 · PubMed
• Schwarzer AC, Aprill CN, Bogduk N. The sacroiliac joint in chronic low back pain. Spine (Phila Pa 1976). 1995;20(1):31–7. doi:10.1097/00007632-199501000-00007 · PubMed
• Maigne JY, Aivaliklis A, Pfefer F. Results of sacroiliac joint double block and value of sacroiliac pain provocation tests in 54 patients with low back pain. Spine (Phila Pa 1976). 1996;21(16):1889–92. doi:10.1097/00007632-199608150-00012 · PubMed
• McCormick ZL, Hurley RW, Anitescu M, et al. Consensus practice guidelines on sacroiliac joint complex pain from a multispecialty, international working group. Pain Med. 2025;26(12):817–917. doi:10.1093/pm/pnaf129 · PubMed