Residency · Residency · Chronic Pain Management
Radiofrequency Ablation of Medial Branch Nerves
Introduction
Radiofrequency ablation (RFA) of the medial branch nerves is one of the most well-established interventional procedures in chronic pain management. By creating a thermal lesion along the nerves that innervate the facet (zygapophyseal) joints, RFA provides durable pain relief for patients with confirmed facetogenic pain. The procedure exists in three modalities -- conventional, cooled, and pulsed radiofrequency -- and understanding the distinctions among them is essential for optimal patient outcomes.
Anatomy and Target Nerves
The medial branch nerves arise from the dorsal rami of the spinal nerves and course over the junction of the transverse process and the superior articular process. Each facet joint receives dual innervation from the medial branch at the same level and the level above. At L5, the dorsal ramus itself (rather than its medial branch) innervates the L5-S1 facet joint and runs in the groove between the sacral ala and the superior articular process of S1. In the cervical spine, the third occipital nerve (TON) supplies the C2-C3 facet joint and requires a specialized technique for ablation.
<image>Detailed anatomical illustration showing a posterior oblique view of the lumbar spine with the medial branch nerves highlighted in yellow as they course over the junction of the transverse process and superior articular process at L3, L4, and L5 levels, with the dorsal ramus at L5 shown crossing the sacral ala. Labels indicate the facet joint capsule, mamilloaccessory ligament, and the dual innervation pattern.</image>
Conventional Radiofrequency Ablation
Conventional RFA uses a continuous current at frequencies of 300-500 kHz to generate temperatures of 60-90 degrees Celsius at the electrode tip. Standard lesion parameters are 80 degrees Celsius for 60-90 seconds per lesion site. The size of the resulting lesion depends on the electrode gauge (typically 18-22 gauge), the active tip length (5-10 mm), and the temperature achieved.
A critical technical detail is that the cannula must be placed parallel to the nerve to maximize the lesion along its course. A perpendicular placement creates only a point lesion that is far more likely to miss the target nerve. Before each lesion is created, sensory stimulation at 50 Hz (threshold less than 0.5 V) confirms proximity to the target nerve, while motor stimulation at 2 Hz (threshold greater than twice the sensory threshold) confirms safety from the ventral ramus.
Cooled Radiofrequency Ablation
Cooled RF uses internally cooled electrodes with water circulating through the cannula tip. The tip temperature is maintained at approximately 60 degrees Celsius while tissue temperatures at the lesion periphery reach 80 degrees or higher. This produces a larger, more spherical lesion compared to conventional RF -- average lesion diameter up to 10 mm versus 5-6 mm for conventional. The larger lesion is particularly advantageous at the sacroiliac joint (lateral branch neurotomy) where nerve location is highly variable. The improved lesion volume may also improve capture rates of target nerves when electrode placement is less precise.
<image>Comparative cross-sectional diagram showing lesion geometry differences between conventional RF (elongated, teardrop-shaped lesion measuring approximately 5-6mm around the active tip), cooled RF (larger, spherical lesion measuring approximately 8-10mm), and pulsed RF (no thermal lesion, showing electromagnetic field distribution around the electrode tip). Temperature gradients illustrated with color mapping from red (hottest) to blue (coolest).</image>
Pulsed Radiofrequency
| Modality | Temperature | Lesion Shape | Lesion Size | Mechanism | Best Use |
|---|---|---|---|---|---|
| Conventional RF | 60–90°C | Elongated (teardrop) | ~5–6 mm | Thermal neurodestructive | Medial branch RFA (standard) |
| Cooled RF | Tip ~60°C; tissue >80°C | Spherical | ~8–10 mm | Thermal with internal cooling | Lateral branch (SI joint); variable nerve course |
| Pulsed RF | <42°C | No thermal lesion | N/A | Electromagnetic neuromodulation | DRG applications; not for facet denervation |
Pulsed RF delivers short bursts of energy (20 ms pulses at 2 Hz) with long silent periods that allow heat dissipation. This keeps tissue temperature below 42 degrees Celsius, avoiding neurodestructive thermal injury. The proposed mechanisms include electromagnetic field effects on neuronal cell membranes, upregulation of c-Fos, and neuromodulation rather than ablation. In head-to-head trials against conventional RFA for facetogenic pain, pulsed RF has shown inferior efficacy. It may have a role in conditions where neurodestructive lesioning carries unacceptable risk, such as dorsal root ganglion applications, but it is not the appropriate modality for facet denervation.
Patient Selection Based on Diagnostic Blocks
Patient selection is the single most important determinant of RFA success. The gold standard is dual comparative diagnostic medial branch blocks, in which the patient undergoes two separate blocks using local anesthetics of different durations (for example, lidocaine and bupivacaine). The International Spine Intervention Society recommends 80% or greater pain relief on both occasions for a positive result. A concordant response means shorter relief with lidocaine and longer relief with bupivacaine, further validating the diagnosis.
Single diagnostic blocks carry a false-positive rate of 25-40%, whereas dual blocks reduce this to approximately 10-15%. The clinical consequence is substantial: patients meeting strict dual-block selection criteria demonstrate success rates of 60-80% after ablation, compared to only 30-50% with single-block paradigms.
Evidence from Randomized Controlled Trials
The evidence base for medial branch RFA is strong when proper patient selection is used. Dreyfuss et al. (2000) demonstrated that 60% of patients achieved at least 90% relief and 87% achieved at least 60% relief at 12 months following lumbar medial branch RFA with strict selection criteria. Nath et al. (2008) conducted a sham-controlled RCT of cervical medial branch RFA showing significant superiority of active treatment at 6 months. Cohen et al. (2010) compared paradigms of 0, 1, and 2 diagnostic blocks before RFA in a multicenter RCT and showed that stricter selection improved success rates but reduced the number of patients eligible for treatment. Van Tilburg et al. (2016) published a systematic review supporting conventional RFA over pulsed RF for facetogenic pain.
Expected Duration of Relief
Typical duration of relief following conventional RFA is 6-12 months. Nerve regeneration occurs at a rate of approximately 1-3 mm/day, and once the nerve regrows, pain may return. Repeat ablation is effective in 75-85% of initial responders, and most patients maintain consistent relief across successive procedures, though some report diminishing returns. Cooled RF may provide slightly longer duration of relief due to its larger lesion size, though comparative data are limited.
<image>Fluoroscopic image illustration showing proper cannula placement for lumbar medial branch radiofrequency ablation at L3, L4, and L5 levels in an oblique view. The cannulae are shown parallel to the medial branch nerves at the junction of the transverse process and superior articular process, with the final cannula at the L5 dorsal ramus positioned at the sacral ala. Bony landmarks including Scotty dog anatomy are labeled.</image>
Complications and Safety
Post-procedure neuritis (a pain flare at the treatment site) occurs in 5-10% of cases and is typically self-limited over 2-6 weeks. Motor nerve injury is rare when proper stimulation thresholds are confirmed before lesioning. Cutaneous numbness or dysesthesia over the treatment area can occur. Infection, hematoma, and allergic reaction are rare. Bilateral procedures at the same level in a single session should be avoided to reduce the risk of paraspinal muscle denervation.
Clinical Pearls
The cannula must always be placed parallel to the target nerve, not perpendicular, to maximize lesion capture along the nerve's course. Sensory and motor stimulation testing should be performed before every lesion to confirm safe and effective cannula placement. Strict patient selection with dual diagnostic blocks and high relief thresholds (80% or greater) is the single most important predictor of RFA success. Cooled RF should be considered when targeting nerves with variable anatomic course, such as the lateral branches at the sacrum. Pre- and post-block pain scores should be meticulously documented, as outcomes data support continued authorization for repeat procedures.
References
- Dreyfuss P, Halbrook B, Pauza K, et al. Efficacy and validity of radiofrequency neurotomy for chronic lumbar zygapophysial joint pain. Spine. 2000;25(10):1270-1277.
- Nath S, Nath CA, Pettersson K. Percutaneous lumbar zygapophysial (facet) joint neurotomy using radiofrequency current, in the management of chronic low back pain: a randomized double-blind trial. Spine. 2008;33(12):1291-1297.
- Cohen SP, Williams KA, Kurihara C, et al. Multicenter, randomized, comparative cost-effectiveness study comparing 0, 1, and 2 diagnostic medial branch (facet joint nerve) block treatment paradigms before lumbar facet radiofrequency denervation. Anesthesiology. 2010;113(2):395-405.
- Van Tilburg CW, Stronks DL, Groeneweg JG, et al. Randomised sham-controlled double-blind multicentre clinical trial to ascertain the effect of percutaneous radiofrequency treatment for lumbar facet joint pain. Bone Joint J. 2016;98-B(11):1526-1533.
