Radiofrequency Treatment in Chronic Pain

Marc Soloman, MD; Mark N Mekhail; Nagy Mekhail, MD, PhD


Expert Rev Neurother. 2010;10(3):469-474. 

In This Article

Clinical Applications

Back pain is the most common complaint in a pain management clinic. This review discusses use of RFA for back pain including facetogenic pain, including cervical and lumbar pain due to sacroiliac (SI) dysfunction, and finally discogenic pain.

Lumbar Facet Nerve Ablation

Lumbar facet nerve ablation is perhaps the most common application of RFA. Lumbar facet arthropathy and pain contributes up to 30% of low back pain in some studies.[14] The superior and inferior articular process from each vertebra forms the facet joint and each joint receives dual innervations from the medial nerve branch of the same level and the level above. Typically the pain is localized in the low back and can radiate to buttocks, hips and thighs, usually above the knee.[15] Cohen et al. found that the only factor associated with a successful outcome was paraspinal tenderness and this should trigger the physician to try facet block.[16]

The general consensus is to start by diagnostic blocks; more than 50% pain relief in more than one block is considered an indication to proceed with RFA. It is well established that lumbar medial branch ablation is an effective means of reducing pain in carefully selected patients on the basis of controlled diagnostic blocks.[17] The use of steroid, sedation and medial nerve branch block (MNBB) versus intra-articular injection for the diagnostic block is debatable.[16,18,19] Patients who undergo cryoablation of the lumber facet medial branches based on positive diagnostic medial branch blocks have better long-term pain relief than patients who are diagnosed by use of per capsular blocks.[18] If strict criteria, including pain relief and ability to perform prior painful movements, are used as the standard for evaluating the effect of controlled local anesthetic blocks, the diagnostic validity of lumbar facet joint nerve blocks may be preserved, even with the use of sedation.[19]

There are currently five randomized trials on the efficacy of RF facet denervation for low back pain. Three of them are positive, one is negative and one is equivocal. In general, one may conclude that with careful patient selection, RF lesioning of the facet joint is effective with a number needed to treat of 1.1 and 1.5.[20]

The procedure is conducted under fluoroscopy and using sterile techniques with the patient in the prone position. The target of the needle for L1–L4 medial branch is the groove formed by the junction of the transverse process and superior articular process. For the L5 dorsal ramus, the target is the notch between superior articular process and sacral ala. To achieve this view, adjust the c-arm in an anteroposterior view with 10–15° oblique to the ipsilateral side and 15–25° to the caudade.[21]

There are two available studies comparing PRF ablation (PRFA) to conventional RFA. The first study, by Kroll et al., concluded that there were no significant differences in pain relief or functional improvement between the randomized groups. However, relative percentage improvement of visual analogue scores (VAS) and Oswestry low back pain disability questionnaire scores at 3 months postprocedure did improve significantly in the conventional RF (CRF) group.[22] In the second study, by Tekin et al., the results showed that VAS and Oswestry Disability Index were lower and maintained in the CRF group at 6 months. At 1 year, the mean VAS score was similar in control and PRF groups but lower in the CRF group than others.[23]

Cervical Facet Nerve Ablation

The cervical facet (zygapophysial) joints are formed by the superior and inferior articular process of two adjacent vertebrae. The joints are innervated by the medial branches of the cervical dorsal rami. Each dorsal ramus sends medial branches to the facet joint at its own level and one level below. The C2–C3 facet joint is innervated by the third occipital nerve (one of the medial branches of the C3 dorsal ramus), and to some degree by one of the medial branches of the C2 dorsal ramus.[24] In patients with chronic pain after whiplash who underwent double-blinded differential block studies (blocking the medial branch), pain in the distribution of the medial branch of the dorsal ramus was confirmed in 54% of patients. The most common levels causing pain have been noted to be C2–C3, followed by C5–C6 and then C4–C5.[25] Paraspinal tenderness is the only clinical variable associated with success. Factors associated with treatment failure included radiation to the head, opioid use and pain exacerbated by neck extension and/or rotation.[26]

From the articular process view, the target is the silhouette between the cervical vertebrae; using a 20 or 22 g, 5 mm active tip, preferably curved, the needle is placed in the target area. From the lateral view the needle is adjusted to lie along the mid-pillar. Sensory stimulation at 50 Hz, 0.3–0.7 V is desirable, followed by motor stimulation at 2 Hz up to 2 V to confirm that there is no motor nerve ablation. Thermal RFA is carried out at 80°C for 90 s.

There are two trials regarding the use of cervical RFA for cervicogenic headaches and both failed to provide solid evidence. Thus, greater occipital nerve block is considered the treatment of choice for cervicogenic headache with infiltration with steroid and local anesthetic. When this procedure is not successful, RF lesioning of the higher facet joints is a treatment option.[20,27,28] When it comes to cervical facet medial nerve RFA, there is one randomized controlled trial by Lord et al. comparing RF facet denervation with a sham intervention for patients suffering chronic cervical pain after a whiplash injury, which showed that 7 months after the intervention the number needed to treat is 1.7 in the active group compared with 12 in the control group.[20,29]

Sacroiliac Joint

The sacroiliac joint complex is a common source of chronic lower back pain. The prevalence among patients with idiopathic low back pain is reported as 18–30%.[30] Patients are selected according to their symptoms, physical exam and diagnostic block. There is no study to compare intra-articular injection of steroid and local anesthetic versus local anesthetic blockade of L5 dorsal rami and lateral branches of S1–S3 nerve root as a method of choice for diagnostic injections. Denervation of L4 and L5 primary dorsal rami and S1–S3 lateral branch RF denervation may provide intermediate-term pain relief and functional benefit in selected patients with suspected SI joint pain. This was reported by Cohen et al. and also by Vallejo et al.[31,32]

Multiple methods have been prescribed to denervate the SI joint, however, we will concentrate on cooled RFA. Kapural et al. found that the majority of patients with chronic SI joint pain experienced a clinically relevant degree of pain relief and improved function following cooled RFA of sacral lateral branches and dorsal rami of L5 at 3–4 months follow-up.[33]

The procedure of cooled RFA is carried out by first identifying the upper sacral foramina by placing three 27-gauge 3.5-inch Quincke needles into the S1, S2 and S3 posterior sacral foramina under x-ray guidance to establish internal reference points. Beginning at the S1 level, an introducer with stylet is inserted onto the bone of the posterior sacrum. The final position of the introducer is 8–10 mm from the lateral edge of point spread function. Once proper electrode location and impedance are achieved, 1 ml of 2% lidocaine is injected and the heating protocol is initiated, delivering RF energy for 2 min and 30 s. Target electrode temperature is 60°C. Once energy delivery is complete, the RF probe is removed. Three lesions are created at each sacral level at 2, 4 and 6 o'clock.

Intervertebral Disc RFA

Pain originating from the intervertebral disc together with the facet joints accounts for over 50% of low back pain. Abnormal noceciptors and unmyelinated nerve fibers, observed in histological studies, in association with degenerative changes, such as annular tears, are most probably the origin of discogenic pain.[34] Multiple modalities have been developed to deliver heat to the annulus of the disc to ablate the nociceptors as well as modify the collagen fibers of the annulus. The intradiscal elector thermal therapy (IDET) aimed to deliver thermal energy at 90°C to the posterior and lateral walls of the disc. IDET literature is rather conflicting. However, in properly selected patients the outcomes have been very favorable.[35] The authors outlined the additional selection criteria, which include positive provocative discography test, maintained disc height (at least 50%), avoidance of multilevel degenerative disc disease (1–2 positive degenerative disc disease at the most), no previous discectomy, no lumbar canal stenosis and no psychological issues that might affect the patient's physical condition.[34,36,37]

Intradiscal biacuplasty aimed to deliver RF energy that heats the posterior annulus at a much lower temperature than IDET. A preliminary report showed significant improvements in patient functional capacity and pain scores, as measured by the VAS, Oswestry Disability Index and MOS Short Form 36.[38] A randomized controlled study is warranted and needed to address the efficacy of the procedure. A cadaver study demonstrated that in the posterior annulus during transdiscal biacuplasty temperatures were adequate for neuroablation. Temperatures reached at the neural foramina and epidural space were low enough to cause neural damage.[39]

The procedure is completed under fluoroscopy; two 17 G transdiscal introducers are placed in the posterior annulus using a posterolateral, oblique approach to gain access to the intervertebral disc. Two RF probes were positioned through each of the introducers bilaterally to create a bipolar RF lesion. The protocol included gradual increase of the temperature to 55°C over 11 min. The final heating interval is 4 min at 55°C. The patient is discharged home with physical therapy instructions and an abdominal binder.

Dorsal Root Ganglion RFA

Dorsal root ganglion is an entity of its own. Partial rhizotomy of the DRG is not intended to treat axial back pain, deafferantion pain or sympathetically maintained pain, such as complex regional pain syndrome (CRPS), but is mainly used for persistent radicular pain. It has been prescribed for the last 30 years. The latest review article about DRG rhizotomy is the excellent work by Malik and Benzon, published in September 2008.[40] The probe is navigated in the neural foramen in the superior dorsal quadrant; sensory stimulation is ideally 50 Hz and 0.5 V, and motor stimulation should be negative at 2 Hz and twice the voltage of the sensory stimulation.

Malik and Benzon reviewed the recent literature (six prospective controlled trials, five retrospective trials), which included studies for cervicogenic headaches, cervical brachial neuropathy and lumbar radiculopathy, as well as studies with PRFA or TRF ablation. The high-quality studies showed that thermal RFA of the DRG is not effective in the treatment of chronic lumbar radicular pain. The trial results provided inconclusive evidence of the efficacy of RFA of DRG in the treatment of cervicogenic headaches. For cervicobrachial pain, there is only short term pain relief with both thermal and pulsed RFA. In general, more studies are still needed to determine the efficacy of DRG ablation.

Sympathetic Ganglia RFA

Sympathetic ganglia RFA is still in use; however most of its application has been replaced, to a great extent, by spinal cord stimulation. This section discusses RFA of the stellate ganglion and lumbar sympathetic chain. Since thermal ablation of the celiac plexus is not feasible, we will describe splanchenic nerve RFA in this section. The practioner usually starts with a diagnostic blockade using local anesthetic before proceeding with RFA. Most of the time the diagnostic blockade and the RFA are carried out on separate occasions; however, in cancer patients the practioner may opt to do both of them at the same time.

The stellate ganglia are responsible for the sympathetic innervation of the face and the upper extremities. Wilkinson published a technique for RF sympatholysis by percutaneous needle placement at T2, T3 and T4. This was refined later as a result of a cadaver study.[41] The success rate ranges from 37–41%.[42] It is indicated for CRPS and ischemic pain. A group reported the use for a patient who experienced residual ischemic symptoms and chronic pain. The patient underwent five stellate ganglion RF neurolysis over a 2-year period, followed by progressive and complete pain relief.[43] Unfortunately, none of these studies are randomized controlled trials. New randomized controlled trials are needed to validate the efficacy and define a measurable and reproducible end point for such procedures.

Celiac plexus block is indicated for upper abdominal pain of visceral origin. The plexus can be ablated chemically by phenol or alcohol. However, another option is to ablate the greater and lesser splanchenic nerves. Splanchenic nerve RFA is usually performed at the T11 or T12 level. It is best described by Raj et al. who reported the result for 22 patients. The procedure was effective for 10 patients with cancer. It was also effective for 12 patients with noncancerous pain, but it had to be repeated in 4 months. There were no complications reported.[44] Raj also described the RF lesions in 100 patients with cancer and chronic pancreatitis with good results.[45] In the prone position, the T12 vertebral body is identified in the posteranterior view of the fluoroscope. From an oblique position (~45°C), the point of entry for both levels is at the junction of the rib and vertebra. From a lateral fluoroscopic view, the needle is advanced until it reaches the junction of anterior one-third and posterior two-thirds of the lateral surface of the vertebral body.[46]

The lumbar sympathetic block is indicated for ischemic cases with suspected sympathetic maintained pain and CRPS pain. Lumbar RF sympatholysis can be conducted at the body of L3 classically or the lower part of L2. Multiple lesions can produce up to 75% relief for 8 weeks.[47] In another study, a 5-mm active tip RF lesion was compared to a phenol block of the lumbar sympathetic chain. The study found that a single phenol block was more effective than multilevel RF lesions. However, phenol neurolysis was associated with a higher incidence of complications.[48]


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: