Radiofrequency (RF) neurotomy is an interventional procedure used to alleviate certain types of low back pain. RF energy is used to thermally coagulate the specific nerves that transmit pain signals. Recent evidence has shown that this procedure demonstrates significant efficacy in relieving low back pain in lumbar zygapophysial joints, and research is ongoing to determine if pain relief for the sacroiliac joint is also possible. This article provides an evidence-based background for performing RF neurotomy, discusses the relevant anatomy, and highlights the indications and technique for lumbar and sacral RF neurotomy.
Radiofrequency neurotomy of the lumbar zygapophysial joints
Percutaneous radiofrequency (RF) medial branch neurotomy involves the use of an RF electrode to ablate the nerves that provide painful sensation in the lumbar zygapophysial (Z)-joints for long-term pain relief. RF neurotomy coagulates a length of the medial branch of a dorsal ramus or a dorsal ramus itself (L5), preventing sensation of the painful Z-joint until the nerve regenerates. An interventionist may use safe, valid, diagnostic medial branch blocks to confirm that low back pain is originating from the Z-joints before using RF neurotomy.
Z-joints are the paired, posterior articulations between vertebrae composed of a synovial capsule, the articular cartilage, and the superior and inferior articular processes surrounded by a joint capsule. The lumbar Z-joints are pain generators with defined referral patterns. Estimates vary regarding the prevalence of Z-joint pain in patients with chronic lower back pain: approximately 6% in a primary care setting, 15% of younger patients, 40% of older patients seen for spine pain ; and 16% of patients with recurrent pain after spine surgery.
A medial branch of a spinal nerve’s dorsal primary ramus innervates the unilateral superior and inferior articular processes of a vertebra. Two unilateral medial branches, the one cephalad and the other caudad to a Z-joint, must be lesioned to provide relief to the Z-joint between them.
Bogduk and colleagues performed a review of key historical and technical RF neurotomy developments, and this subsection refers specifically to percutaneous RF neurotomy as parallel placement of an RF electrode adjacent to a nerve supplying sensation to a Z-joint to achieve thermal ablation. The targeted nerve segment is the same in a diagnostic anesthetic block and a percutaneous RF neurotomy.
Background of RF neurotomy
Medial branch neurotomy is a valid, safe, effective technique for long-term relief of pain to Z-joints. Earlier techniques, such as rhizolysis with scalpel or facet denervation with RF electrodes, reportedly provided pain relief but lacked concept validity. It was not anatomically possible with either technique to lesion the medial branch nerves, the articular branches, or the L5 dorsal ramus that innervate the Z-joints. Medial branch neurotomy emerged as an anatomically valid modification with the important technical aspect of parallel orientation of the tip of the RF electrode to the medial branch to reliably lesion the maximal length of this nerve. Peripheral nerves regenerate in a proximal-to-distal direction at a certain distance per unit of time. Therefore, the longer the nerve that is completely lesioned, the longer the duration of denervation and pain relief. Dreyfuss and colleagues (in 2000) demonstrated sustained effective relief of chronic lumbar Z-joint pain in a prospective study with patients selected after controlled diagnostic blocks. Lumbar RF neurotomy has an average duration of relief of 10.5 months and can be safely repeated with equal duration of relief and efficacy.
To conduct RF neurotomy, a physician needs an alternating current voltage generator connected to the active and reference electrodes, an active electrode positioned approximately parallel to the nerve and a reference electrode, and an adhesive pad with a large surface area applied to the patient’s skin. The RF tissue lesion is produced in the immediate vicinity of the active electrode needle tip. The lesion extends approximately 1 to 2 electrode widths, a maximal effectiveness of 2 mm, away from the active needle tip in the shape of an oblate spheroid. The active electrode needle is typically an insulated 22-gauge Sluijter-Mehta cannula 5, 10, or 15 cm long with an exposed metal tip to permit current flow. The rapidly alternating current fields oscillate charged molecules. Tissue electrical impedance to current flow produces heat in the tissue around the active electrode needle tip, which coagulates and destroys unmyelinated and myelinated fibers in the nearby nerve. An oblate spheroid lesion is produced around the exposed needle. A physician can enlarge the size of the lesion by increasing the needle size (22, 20, or 18 gauge), exposed metal tip length (5, 10, or 15 mm), lesioning time (60 to 90 seconds), or number of needle placements or by repeated lesioning.
Background of RF neurotomy
Medial branch neurotomy is a valid, safe, effective technique for long-term relief of pain to Z-joints. Earlier techniques, such as rhizolysis with scalpel or facet denervation with RF electrodes, reportedly provided pain relief but lacked concept validity. It was not anatomically possible with either technique to lesion the medial branch nerves, the articular branches, or the L5 dorsal ramus that innervate the Z-joints. Medial branch neurotomy emerged as an anatomically valid modification with the important technical aspect of parallel orientation of the tip of the RF electrode to the medial branch to reliably lesion the maximal length of this nerve. Peripheral nerves regenerate in a proximal-to-distal direction at a certain distance per unit of time. Therefore, the longer the nerve that is completely lesioned, the longer the duration of denervation and pain relief. Dreyfuss and colleagues (in 2000) demonstrated sustained effective relief of chronic lumbar Z-joint pain in a prospective study with patients selected after controlled diagnostic blocks. Lumbar RF neurotomy has an average duration of relief of 10.5 months and can be safely repeated with equal duration of relief and efficacy.
To conduct RF neurotomy, a physician needs an alternating current voltage generator connected to the active and reference electrodes, an active electrode positioned approximately parallel to the nerve and a reference electrode, and an adhesive pad with a large surface area applied to the patient’s skin. The RF tissue lesion is produced in the immediate vicinity of the active electrode needle tip. The lesion extends approximately 1 to 2 electrode widths, a maximal effectiveness of 2 mm, away from the active needle tip in the shape of an oblate spheroid. The active electrode needle is typically an insulated 22-gauge Sluijter-Mehta cannula 5, 10, or 15 cm long with an exposed metal tip to permit current flow. The rapidly alternating current fields oscillate charged molecules. Tissue electrical impedance to current flow produces heat in the tissue around the active electrode needle tip, which coagulates and destroys unmyelinated and myelinated fibers in the nearby nerve. An oblate spheroid lesion is produced around the exposed needle. A physician can enlarge the size of the lesion by increasing the needle size (22, 20, or 18 gauge), exposed metal tip length (5, 10, or 15 mm), lesioning time (60 to 90 seconds), or number of needle placements or by repeated lesioning.
Anatomy
A lumbar spinal nerve root gives rise to a dorsal and ventral primary ramus just lateral to the intervertebral foramen. The dorsal ramus splits into a lateral and medial branch. The lateral branch innervates the longissimus and iliocostalis muscles and the overlying skin in the lower back. The medial branch innervates a superior articular process, an inferior articular process, the periosteum of the vertebral arch, the multifidus and interspinous muscles, and the interspinous ligament. The lumbar medial branch presents an anatomically consistent target for neurotomy, because it pierces the intervertebral ligament dorsally and runs inferomedially in a groove between the superior articular process and the root of the transverse process until covered by the mamillo-accessory ligament. Distal to the mamillo-accessory ligament, the multiple divisions of the medial branch are small and variable, making the specific division to the articular process a difficult target.
The L5 dorsal ramus itself is ablated for lesioning of the L5-S1 Z-joint, along with the L4 medial branch.
Indications, contraindications, and complications
The patient should have chronic back pain that is unresponsive to conservative care but has responded positively to controlled diagnostic blocks. Controlled diagnostic blocks of the nerve supply (medial branch, L5 dorsal ramus) to the Z-joints must produce significant, temporally concordant pain relief and functional gains in pain-limited activities before performing percutaneous RF neurotomy. Clinicians’ views vary on thresholds indicating significant pain relief. Dreyfuss and colleagues used a threshold of at least 80% relief, which is considered the gold standard. A history of spine surgery or prior RF neurotomy is not a contraindication; however, hardware or ossified scar tissue may prevent adequate access to the target nerve.
Generally, indications and contraindications of RF neurotomy are similar to those of the diagnostic medial branch blocks completed first (refer to article on diagnostic medial branch blocks by Dr Kennedy). RF is safe, with potential for standard percutaneous complications of infection or bleeding. A current infection, a history of bleeding, or a coagulopathy are contraindications. Localized back pain has been the only commonly observed complication in about 1% of patients. Short-term painful neuritis for a few weeks is an uncommon complication.
Dreyfuss and colleagues studied 5 patients with lumbar spine magnetic resonance imaging (MRI) at an average of 21 months after medial branch neurotomy to determine the effects of deep paraspinal muscle denervation. These patients had diffuse multifidus atrophy that could not be reliably localized to the ablated levels by MRI, and patients had ongoing pain relief with no complications. Electrodiagnostic study may explain this phenomenon of diffuse atrophy because “lumbar multifidus muscle is polysegmentally innervated.” Therefore, multifidus muscle continues to support the lumbar spine despite selective denervation of the levels indicated by medial branch blocks.
Technique
Key elements of the procedure technique are discussed in this section, with greater detail available in the reference texts. The patient must be educated regarding the technique preprocedure, so that sensory and motor stimulation testing can be completed at the physician’s discretion. The patient is placed in a prone position, sterile technique is used, and vital signs are monitored throughout the procedure. The reference electrode pad is applied to the patient’s skin and attached to the RF generator. A sterile active RF electrode and cable is attached to the RF generator. A 3.5-in, 25-gauge spinal needle is used with local anesthetic to infiltrate the trajectory of the larger-gauge RF needle. The area closest to the target nerve is not anesthetized if the physician is to perform sensory and motor stimulation testing. Typically, intermittent fluoroscopy is used to guide the RF needle on a trajectory to achieve optimal parallel alignment of the target nerve and the RF needle tip. Needle tip placement is confirmed in anteroposterior, oblique, and lateral views. Some practitioners confirm proximity to the target nerve with the presence of a sensory response, and sufficient distance from the ventral ramus with the absence of peripheral motor activation or radicular pain. A positive sensory response is indicated by a tingling or pressure felt by the patient with 50 Hz stimulation at less than 0.5 V. A reassuring motor check is the absence of radicular pain and activation of muscles in the corresponding myotomal distribution at 2 Hz up to 3 V. If RF needle tip placement is satisfactory, a local anesthetic is injected through the RF needle cannula to prevent the sensation of the ablation. Typically, the ablation is conducted for 60 to 120 seconds at 80°C. The process of active electrode needle placement is repeated to ablate both nerves for each Z-joint. The needle is withdrawn and sterile dressings are applied. The patient is monitored after the procedure for any complications and is discharged with follow-up information.