Facet Joint Interventions · Evidence-Based Patient Guide
Facet Joint Rhizotomy · 34 References

Facet joint rhizotomy and injections: 10 questions every patient should ask.

A well-established technique with solid evidence, three very different technologies, and one selection step that determines everything: the diagnostic block.

Published July 2026 • By Dr. Rodrigo Ávila Cervantes • Spine Neurosurgeon · FAANS · FCNS • 34 references · ~18 min read

The facet joints — also called zygapophysial joints or simply “facets” — are the paired synovial joints that connect adjacent vertebrae through their posterior elements. Each vertebra has four of them (two superior, two inferior), for a total of 96 facet joints in the entire spine. They are lined with hyaline cartilage, surrounded by a fibrous capsule, and lubricated with synovial fluid — structurally similar to a knee or hip joint, just smaller. They bear between 15% and 25% of axial compressive load in the lumbar spine, and this percentage increases significantly with extension and rotation. (Kapetanakis, 2021; Jaumard, 2011)

Pain from the facet joints is generated through several mechanisms: capsular stretch and distension, inflammatory mediators released during articular degeneration, and direct activation of the nociceptive nerve fibers that innervate the joint capsule — substance P and calcitonin gene-related peptide (CGRP) have both been identified in facet joint capsule nerve terminals. The nerve responsible for transmitting this pain is the medial branch of the dorsal ramus — a small nerve that crosses the transverse process at each lumbar level. This nerve is the anatomical target of both diagnostic blocks and rhizotomy. (Cavanaugh, 1996; McLain, 1998)

15–40%
of chronic low back pain is estimated to be facetogenic in origin (Manchikanti 2020)
54–67%
of chronic neck pain has a zygapophysial joint component — especially after whiplash (Lord 1996, Barnsley 1994)
96
total facet joints in the human spine — any of them can become a pain generator

This is the most important diagnostic question in facet joint medicine — and the honest answer is that clinical examination and imaging alone cannot reliably distinguish facet joint pain from discogenic pain. There is no physical examination finding that is specific to facet joint pain. No imaging finding confirms that a visibly arthritic facet joint is the source of a particular patient’s symptoms. Facet arthropathy on MRI or CT is common in asymptomatic people. (Schwarzer, 1994; Kwee, 2021)

Clinical features that suggest facetogenic pain — not diagnostic, but suggestive
  • ✓Predominantly axial pain (back or neck) — not pain that primarily radiates into the leg or arm
  • ✓Worse with extension and rotation of the spine (movements that load the facet joint); relieved by flexion
  • ✓Referral into the buttock or thigh (not below the knee) in lumbar involvement; into the shoulder or arm in cervical
  • ✓Worse in the morning and after prolonged sitting or standing; briefly relieved by movement
  • ✗The absence of neurological deficit — numbness, weakness, or reflex change below the knee strongly suggest a radicular (nerve root) cause, not facet joint

The only reliable method for confirming that facet joint pain is the source of a patient’s symptoms is a controlled diagnostic nerve block that temporarily abolishes the pain. This is not a therapeutic injection — it is a diagnostic test. The diagnostic algorithm has three steps: (1) clinical suspicion based on history and pattern, (2) first confirmatory medial branch block, (3) second confirmatory medial branch block with a different-duration anesthetic on a separate occasion. (Dreyfuss, 2002; Said, 2023)

Why imaging is insufficient for diagnosis

Facet arthropathy on MRI or CT correlates poorly with pain. Multiple studies have confirmed that the degree of radiographic degeneration does not predict who has facet joint pain and who does not. A patient with severe facet arthropathy on imaging may have entirely discogenic pain; a patient with minimal imaging changes may have predominantly facetogenic pain. Imaging identifies anatomy. The diagnostic block identifies physiology.

The medial branch of the dorsal ramus is a small nerve that crosses the transverse process at each lumbar level before entering the facet joint capsule. Each lumbar facet joint receives innervation from two medial branches — one from the level above and one from the same level. To block a single facet joint, both medial branches must be blocked. A small volume of local anesthetic (typically 0.5–1.0 mL) is deposited at each medial branch under fluoroscopic guidance, in contact with the nerve. (Bogduk, 2009; Dreyfuss, 2002)

Why two blocks on separate days? The false-positive rate of a single medial branch block is 25–40%. This means that 1 in 3 patients who report significant pain relief after a single block would not benefit from rhizotomy — the relief was a placebo response or nonspecific effect. The two-block protocol with different-duration anesthetics (lidocaine for the first, bupivacaine for the second, or vice versa) requires that the patient reports relief that is concordant with the expected duration of each agent. When both blocks are positive and the durations of relief are concordant, the false-positive rate drops below 5%. This is the single most important factor in the outcome of rhizotomy. (Lord, 1996; Dreyfuss, 2002)

An intraarticular facet joint injection places corticosteroid directly inside the joint space under fluoroscopic or CT guidance. The goal is to reduce intraarticular inflammation. This is mechanistically different from a medial branch block (which targets the nerve outside the joint) and from rhizotomy (which ablates the nerve that transmits pain from the joint). The distinction matters because the evidence for each is different. (Said, 2023; Carette, 1991)

What the NEJM trials showed about intraarticular injections

Carette et al. (N Engl J Med, 1991) randomized 101 patients with chronic low back pain to intraarticular facet joint methylprednisolone versus saline. At 1 month and 6 months, no significant difference in pain or disability between the two groups. Barnsley et al. (N Engl J Med, 1994) found the same result for chronic cervical facet pain after whiplash. These two NEJM trials established that intraarticular corticosteroid injections do not produce meaningful long-term benefit for facet joint pain — they may offer modest short-term relief (4–6 weeks) for an acute inflammatory component, but they are not a definitive treatment. (Carette, 1991; Barnsley, 1994)

Conventional radiofrequency (RF) neurotomy — also called medial branch neurotomy or facet denervation — is a minimally invasive procedure that uses radiofrequency energy to produce thermal coagulation of the medial branch nerves, permanently interrupting the pain signal from the facet joint. The electrode tip is heated to 80–90°C for 90 seconds, producing a controlled lesion around the nerve. The procedure is performed under fluoroscopic guidance with the electrode positioned parallel to the course of the medial branch nerve — parallel positioning is critical because the lesion is cylindrical and the nerve must pass through it to be ablated. (Bogduk, 2009; Said, 2023)

The foundational evidence — what the trials actually showed
  • 1Lord et al. N Engl J Med 1996 — The foundational RCT. 24 patients with chronic cervical zygapophysial joint pain confirmed by two diagnostic blocks, randomized to RF neurotomy vs. sham. Median duration of pain relief: 263 days (RF) vs. 8 days (sham). The most cited trial in facet joint RF literature.
  • 2Dreyfuss et al. Spine 2000 — Lumbar RF. 60% of patients achieved ≥90% pain relief at 12 months after properly performed RF neurotomy confirmed by two diagnostic blocks.
  • 3MacVicar et al. Pain Med 2013 — Large real-world cohort confirming similar results in clinical practice when two-block selection protocol is rigorously applied.
  • 4Juch et al. JAMA 2017 (MINT trials) — Three concurrent RCTs. No significant difference between RF + usual care vs. usual care alone at 3, 6, 12 months. Critical methodological flaw: only one diagnostic block required, not two — admitted a false-positive rate that diluted the treatment group with non-responders. When only one block is used, up to 40% of the ‘positive’ group has no facet joint pain. This trial’s null result likely reflects patient selection failure, not treatment failure.

Three technologies are currently used or marketed for facet joint denervation: conventional radiofrequency, pulsed radiofrequency, and laser. They are often presented as equivalent options, or the laser is presented as a “more advanced” technology. The evidence tells a different story.

Standard of care
Conventional Radiofrequency (thermal)
Mechanism: Thermal coagulation of the medial branch nerve at 80–90°C × 90 seconds. Produces irreversible necrosis of the nerve within the lesion zone.

Evidence: Level I — 2 NEJM RCTs, multiple high-quality RCTs and systematic reviews. The most rigorously studied facet intervention.

Duration: 6–18 months until nerve regeneration.

Best for: Lumbar and cervical facet joint pain confirmed by two diagnostic blocks.
Evidence-based standard
Different indication
Pulsed Radiofrequency (PRF)
Mechanism: Pulsed delivery of RF energy (20ms on / 480ms off) — temperature kept below 42°C. NOT thermal destruction. Modulates pain processing through electrical field effects without destroying the nerve.

Evidence: Level III-IV — weaker than conventional RF for medial branch territory. Tekin 2007 showed conventional RF superior to PRF at 6 months.

Where it has a role: Dorsal root ganglion modulation (DRG-PRF); where nerve destruction would be harmful. Not the correct tool for classical medial branch rhizotomy.
Useful, but different indication
Evidence gap
Laser Rhizotomy
Mechanism: Photothermal energy (Nd:YAG, diode, CO₂) delivered via percutaneous or endoscopic approach. Produces tissue ablation through heat, similar end goal to conventional RF but different delivery system.

Evidence: No published randomized controlled trial comparing laser rhizotomy to conventional RF for facet joint pain under controlled conditions. Limited case series suggest comparable short-term outcomes to RF.

The problem: Higher cost and technical complexity without demonstrated superiority. The bar for replacing an evidence-based standard is a controlled trial — that trial does not exist for laser facet rhizotomy.
No RCT vs. conventional RF
The bottom line on laser vs. RF

Laser rhizotomy and conventional RF rhizotomy share the same therapeutic goal: thermal ablation of the medial branch nerve. The difference is the energy delivery system. In the absence of a randomized controlled trial showing that laser produces superior outcomes to conventional RF, the evidence hierarchy requires conventional RF as the standard. A patient offered laser rhizotomy should ask: what randomized controlled trial shows this is better than conventional radiofrequency? As of 2026, that trial does not exist. (Said, 2023; Poetscher, 2014; Leggett, 2014)

The medial branch nerve always regenerates. This is not a complication — it is the expected biological course of peripheral nerve axon regrowth after thermal ablation. Regeneration typically occurs between 6 and 18 months after the procedure. Pain returns as the nerve recovers innervation of the facet joint. This is why patients often report that “the rhizotomy stopped working” at a predictable interval. (North, 1994; MacVicar, 2013)

The practical implication is important: rhizotomy is a repeatable procedure. When pain returns after a successful RF rhizotomy, the procedure can be repeated with the expectation of similar results, without the need to repeat the two-block diagnostic protocol (provided the original diagnosis remains valid and the pain pattern has returned to its pre-rhizotomy character). Multiple repetitions are documented in the literature without loss of efficacy over successive cycles. (MacVicar, 2013; North, 1994)

Ideal candidate for facet joint rhizotomy
  • ✓Two positive comparative diagnostic blocks producing at least 80% pain relief on both occasions, with concordant duration (shorter relief with lidocaine, longer with bupivacaine)
  • ✓Predominantly axial pain, worse with extension and rotation, without significant radicular component (pain below the knee or below the elbow)
  • ✓Failure of at least 3–6 months of appropriate conservative treatment
  • ✓No active infection, no coagulopathy, no pregnancy
Who should NOT have rhizotomy — or should be approached with caution

Patients with only ONE diagnostic block (not two) — too high a false-positive rate to justify the procedure. Patients with significant radicular pain (pain, weakness, or numbness below the knee) — this is a nerve root problem, not a facet joint problem; rhizotomy does not treat radiculopathy. Patients with significant psychosocial factors (depression, catastrophizing, active litigation, somatization) — these are the strongest predictors of a poor outcome after any spine procedure, including rhizotomy. Patients with active spinal infection. Patients on anticoagulation without a safe bridge protocol. (North, 1994; Cohen, 2007)

The principle is the same throughout the spine — thermal ablation of the medial branch nerve that innervates the facet joint — but the anatomical targets and technical execution differ significantly by region. (Manchikanti, 2016; Cohen, 2007)

Region Common levels Target nerve Evidence level Key notes
Lumbar L3-L4, L4-L5, L5-S1 Medial branch of dorsal ramus (L3-L4); L5 dorsal ramus for L5-S1 joint Highest — most RCTs L5-S1 requires targeting the L5 dorsal ramus at the sacral ala — different landmark
Cervical C3-C7; Third occipital nerve for C2-C3 Cervical medial branches; third occipital nerve for C2-C3 High — NEJM trial (Lord 1996) C2-C3 joint has unique innervation via third occipital nerve — requires specific technique
Thoracic T3-T10 (less common) Thoracic medial branches Lowest — least studied Technically more challenging; rib and costotransverse joints complicate access; less common indication

This question has three distinct clinical meanings, and the answer is different for each. Understanding the distinction is important before agreeing to a treatment plan that involves multiple sessions.

Three meanings of ‘multiple radiofrequency procedures’ — and what the evidence says for each
  • 1Repeat procedures at the same level over time (nerve regenerated, pain returned, RF repeated): NOT additive. Each cycle produces similar relief to the previous one — the nerve regenerates to its baseline, and the next RF procedure starts from zero. MacVicar et al. (2013) confirmed consistent outcomes across multiple repeat cycles without evidence of cumulative benefit or diminishing returns. This is good news: the procedure does not ‘wear out’ with repetition.
  • 2Treating multiple levels simultaneously (e.g., L3-L4, L4-L5, and L5-S1 confirmed by diagnostic blocks and all treated in the same session): YES, in a practical sense. Each level contributes independently to the pain burden. Denervating three confirmed levels in one session addresses three separate pain generators — the cumulative clinical improvement can be substantial. This is standard practice and is not ‘additive’ in a pharmacological sense, but rather additive in that more sources are addressed. The prerequisite: each level must be individually confirmed by diagnostic blocks. Treating additional levels without diagnostic confirmation is not appropriate.
  • 3More frequent procedures to extend the effect (doing RF before the nerve has fully regenerated, hoping to ‘build up’ the effect): NO evidence of benefit, and not recommended. The nerve regenerates on its own biological timeline (6–18 months) regardless of how many times it has been ablated previously. Performing RF before pain has returned provides no additional benefit and exposes the patient to unnecessary procedural risk. The correct trigger for repeat RF is the return of the original pain pattern to its pre-RF character, not a calendar interval.
Can RF rhizotomy and intraarticular injections be combined?

Yes — and this combination has a logical mechanistic rationale. Intraarticular steroid injections target the inflammatory component of facet joint pain (short-term, typically 4–12 weeks). RF rhizotomy targets the pain signal transmission pathway (longer-term, 6–18 months). They act on different parts of the pain mechanism and are not redundant. In practice: an acute severe flare may be managed with an intraarticular injection to allow the patient to tolerate the diagnostic block process; once two blocks are positive, RF provides the longer-term solution. Using intraarticular injections repeatedly as a substitute for RF in a patient who is an appropriate RF candidate, however, is not supported by the NEJM evidence — the injections do not produce meaningful long-term benefit in isolation. (Carette, 1991; MacVicar, 2013; Bogduk, 2009)

— Facet joint rhizotomy is one of the best-supported minimally invasive spine procedures. The key is what happens before the procedure, not during it.

Conventional radiofrequency medial branch neurotomy is one of the few minimally invasive spine procedures with Level I evidence from randomized controlled trials published in the New England Journal of Medicine. When performed in properly selected patients — those confirmed by two comparative diagnostic blocks — it produces meaningful, sustained pain relief in 60–80% of patients, with a well-documented safety profile. The nerve always grows back, the procedure is repeatable, and outcomes are consistent across successive cycles. This is an established, accepted technique.

The MINT trial (JAMA 2017), which showed no significant difference versus usual care, is the most cited criticism of RF rhizotomy. But the MINT trial had a fundamental selection flaw: it required only one diagnostic block, not two — knowingly accepting a 25–40% false-positive rate into the treatment group. When up to 40% of the treated patients did not actually have facet joint pain, a null result is not surprising. The trial measured the failure of patient selection as if it were the failure of the treatment. Studies that use the two-block protocol consistently show the opposite result.

On the question of laser versus conventional RF: both achieve the same therapeutic goal by the same mechanism — thermal ablation of the medial branch nerve. In the absence of a randomized controlled trial demonstrating that laser produces superior outcomes to conventional RF for facet joint pain, there is no evidence-based justification for preferring laser over RF. The higher cost of laser does not translate to higher evidence. A patient offered any non-RF technique for facet rhizotomy should ask for the controlled trial that supports that choice.

The most important factor in the outcome of any facet joint procedure is not the technology used — it is the rigor of patient selection. Two comparative diagnostic blocks, properly performed, with concordant responses: this is what determines whether a rhizotomy will work. The best RF system in the world cannot help a patient who does not have facet joint pain.

Dr. Rodrigo Ávila Cervantes
Spine Neurosurgeon · FAANS · FCNS · CICOVE Director · Hospital Ángeles del Pedregal
CICOVE · Hospital Ángeles del Pedregal · Mexico City

Were you offered a facet procedure? Send us your imaging and block records.

Dr. Ávila reviews each case personally. Response within 48–72 hours. No charge for the case review.

Free pre-evaluation WhatsApp · Dr. Ávila
Related article
Conservative treatment for spine pain: what works, what doesn't, and what carries risk
Read →

References

34 peer-reviewed references · 1991–2023 · Including 2 NEJM RCTs and systematic reviews from Spine, Pain Medicine, Anesthesiology, and JAMA

  1. Kapetanakis S, Gkantsinikoudis N. Anatomy of lumbar facet joint: a comprehensive review. Folia Morphol (Warsz). 2021;80(4):799-805. PMID: 33084010
  2. Jaumard NV, Welch WC, Winkelstein BA. Spinal facet joint biomechanics and mechanotransduction in normal, injury and degenerative conditions. J Biomech Eng. 2011;133(7):071010. PMID: 21823749
  3. Kalichman L, Hunter DJ. Lumbar facet joint osteoarthritis: a review. Semin Arthritis Rheum. 2007;37(2):69-80. PMID: 17379279
  4. Cavanaugh JM et al. Lumbar Facet Pain: Biomechanics, Neuroanatomy and Neurophysiology. J Biomech. 1996;29(9):1117-1129. PMID: 8923288
  5. McLain RF, Pickar JG. Mechanoreceptor endings in human thoracic and lumbar facet joints. Spine. 1998;23(2):168-173. PMID: 9474726
  6. Kwee RM, Kwee TC. Imaging of facet joint diseases. Clin Imaging. 2021;80:167-179. PMID: 34333352
  7. Anaya JEC et al. Differential Diagnosis of Facet Joint Disorders. Radiographics. 2021;41(2):543-558. PMID: 33481690
  8. Schwarzer AC et al. Clinical features of patients with pain stemming from the lumbar zygapophysial joints. Spine. 1994;19(10):1132-1137. PMID: 8059258
  9. Manchikanti L et al. Utilization of Facet Joint and Sacroiliac Joint Interventions in Medicare Population from 2000 to 2014. Curr Pain Headache Rep. 2016;20(10):58. PMID: 27646014
  10. Manchikanti L et al. Utilization Patterns of Facet Joint Interventions: Retrospective Cohort Study. Curr Pain Headache Rep. 2019;23(10):73. PMID: 31388874
  11. Dötterl A. Don't Forget the Facet Joint Syndrome. Dtsch Arztebl Int. 2023;120(20):364. PMID: 37493051
  12. Said N, Amrhein TJ, Joshi AB et al. Facets of facet joint interventions. Skeletal Radiol. 2023;52(10):1873-1886. PMID: 36245007
  13. Lord SM et al. Chronic cervical zygapophysial joint pain after whiplash: a placebo-controlled prevalence study. Spine. 1996;21(15):1737-1744. PMID: 8839540
  14. Dreyfuss P et al. Specificity of lumbar medial branch and L5 dorsal ramus blocks. Spine. 2002;27(24):2904-2913. PMID: 11935105
  15. Manchikanti L et al. Cervical zygapophysial (facet) joint pain: effectiveness of interventional management strategies. Postgrad Med. 2016;128(1):54-68. PMID: 26653406
  16. Schwarzer AC et al. The prevalence and clinical features of internal disc disruption in patients with chronic low back pain. Spine. 1995;20(17):1878-1883. PMID: 7570623
  17. Carette S et al. A controlled trial of corticosteroid injections into facet joints for chronic low back pain. N Engl J Med. 1991;325(14):1002-1007. PMID: 1861420
  18. Barnsley L et al. Lack of effect of intraarticular corticosteroids for chronic pain in the cervical zygapophyseal joints. N Engl J Med. 1994;330(15):1047-1050. PMID: 8127340
  19. Mazmudar A, Nayak R, Patel AA. Therapeutic Facet Joint Interventions in the Lumbar Spine: An Economic Value Perspective. Clin Spine Surg. 2020;33(10):411-417. PMID: 32657844
  20. Manchikanti L et al. Comprehensive evidence-based guidelines for facet joint interventions in the management of chronic spinal pain: ASIPP guidelines 2020. Pain Physician. 2020;23(3S):S1-S127. PMID: 32544166
  21. Lord SM et al. Percutaneous radio-frequency neurotomy for chronic cervical zygapophysial-joint pain. N Engl J Med. 1996;335(23):1721-1726. PMID: 8551870
  22. Dreyfuss P et al. Efficacy and validity of radiofrequency neurotomy for chronic lumbar zygapophysial joint pain. Spine. 2000;25(10):1270-1277. PMID: 10806003
  23. van Kleef M et al. Randomized trial of radiofrequency lumbar facet denervation for chronic low back pain. Spine. 1999;24(18):1937-1942. PMID: 10515014
  24. MacVicar J et al. Lumbar medial branch radiofrequency neurotomy in New Zealand. Pain Med. 2013;14(5):639-645. PMID: 23521608
  25. Cohen SP, Raja SN. Pathogenesis, diagnosis, and treatment of lumbar zygapophysial (facet) joint pain. Anesthesiology. 2007;106(3):591-614. PMID: 17325518
  26. Bogduk N, Dreyfuss P, Govind J. A narrative review of lumbar medial branch neurotomy for the treatment of back pain. Pain Med. 2009;10(6):1035-1045. PMID: 19622104
  27. Poetscher AW et al. Radiofrequency denervation for facet joint low back pain: a systematic review. Spine. 2014;39(14):E842-E850. PMID: 24732843
  28. Geurts JW et al. Efficacy of radiofrequency procedures for spinal pain: a systematic review of RCTs. Reg Anesth Pain Med. 2001;26(5):394-400. PMID: 11561256
  29. Juch JNS 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. PMID: 28672319
  30. Tekin I et al. A comparison of conventional and pulsed radiofrequency denervation in the treatment of chronic facet joint pain. Clin J Pain. 2007;23(6):524-529. PMID: 17575495
  31. Chua NHL et al. Pulsed radiofrequency treatment in interventional pain management: mechanisms and potential indications. Acta Neurochir. 2011;153(4):763-771. PMID: 21191611
  32. Facchini G et al. A comprehensive review of pulsed radiofrequency in the treatment of pain associated with different spinal conditions. Br J Radiol. 2017;90(1073):20150406. PMID: 27603683
  33. North RB et al. Radiofrequency lumbar facet denervation: analysis of prognostic factors. Pain. 1994;57(1):77-83. PMID: 8065800
  34. Leggett LE et al. Radiofrequency ablation for chronic low back pain: a systematic review of randomized controlled trials. Pain Res Manag. 2014;19(5):e146-e153. PMID: 25136579
Dr. Rodrigo Ávila Cervantes · CICOVE · Hospital Ángeles del Pedregal · Ciudad de México
Privacy Notice  ·  dravilaspine.us  ·  cicovespine.us  ·  © 2026