Radiofrequency neurotomy interrupts the nociceptive afferent from the Z-joint by thermally coagulating the two medial branches that innervate a given Z-joint. The exposed terminal portion of radiofrequency probe delivers heat at 80°C. For each Z-joint (except the C2-3 joint, which is innervated by the third occipital nerve), two medial branches need to be ablated.
Radiofrequency neurotomy can provide relative long-term benefit symptoms from persistent or recurrent Z-joint pain despite conservative care (that have had transient benefit from Z-joint injections), and for patients with substantial (e.g., at least 80%) pain relief after dual blocks of the medial branch with two local anesthetics of different duration on two occasions.
Lumbar Medial Branch Neurotomy
Since the “groove” at the junction of the superior articular process and the transverse process can be clearly viewed in a certain fluoroscopic position, this “groove view” has been proposed for the starting position of a lumbar medial branch neurotomy. Specifically, the C-arm intensifier is obliqued ipsilaterally approximately 10 to 15 degrees and tilted cephalad 20 degrees. A 22-gauge 100 mm radiofrequency probe with a 5-mm active tip is then introduced using the “tunnel view” until it contacts the bone and advanced along the “groove” where the medial branch resides. The radiofrequency probe creates an effective circumferential lesion around the probe but does so poorly distal to the tip. To avoid incomplete heating, the radiofrequency probe needs to be placed on and parallel to the path of the medial branches crossing the transverse processes. In AP imaging, the needle should be medial to the lateral silhouette of the superior articular process. A lateral view should be obtained to ensure that the needle is not positioned anterior to the posterior aspect of the neuroforamen. Motor stimulation using 2.0 Hz and less than 2.0 V should not induce any muscle twitching or movement in the lower extremity. A single lesion at 80 degrees celsius for 60 seconds is performed. A second and third lesion is performed after repositioning the needle 1.0 mL cephalad and caudad to ensure proper coagulation of the length of the medial branch.
Cervical Medial Branch Neurotomy
For cervical medial branch ablation, the patient is placed in a prone position and the C-arm is tilted in a cephalad direction to obtain a pillar view. The first radiofrequency lesioning probe is inserted parallel to the articular pillar, directed down the fluoroscopic beam and slightly medial until it touches the dorsal aspect of the pillar. The needle is then walked lateral until it slips off the bone. A second radiofrequency probe is inserted with approximately 30 degrees of ipsilateral obliquity and slightly caudad so that this needle can be positioned more anteriorly. The rest of the needle advancement is the same as for the first needle. This dual needle placement positions the two probes to allow subsequent adequate denervation along the entire length of the medial branch. Prior to lesioning, a lateral view is taken to ensure that the tips of the radiofrequency probes are not anterior to the anterior margin of the articular pillars. Motor stimulation is conducted at a frequency of 2.0 Hz and at less than 2.0 V intensity. If muscle twitching or upper extremity movement occurs, this indicates that the radiofrequency probe is too close to the anterior rami and needs to be repositioned. Next, 0.5 mL of 1% lidocaine is injected through the radiofrequency cannula. A lesion is then performed at this position at 80°C for 60 seconds. The probe is then repositioned 1 mm caudad and cephalad, respectively, and two additional lesions are made for a total of six lesions using the two radiofrequency probes. A similar technique is employed for the third occipital nerve.
Proper patient selection is the key for optimal outcome from radiofrequency neurotomy. One recent prospective study has demonstrated good efficacy from the procedures when patients with presumed Z-joint pain are selected using the double block paradigm with comparative local anesthetic (130). The patients were diagnosed with a lumbar Z-joint pain if they obtained at least 80% pain reduction after medial branch blocks with 0.5 mL of 2% lidocaine on one occasion and 0.5% bupivicaine on another. At 12 months following the radiofrequency medial branch neurotomy, 60% of patients achieved at least 90% pain reduction and 87% of patients had 60% pain relief. The success of denervation was seen in virtually all patients as demonstrated by post-radiofrequency needle EMG of the corresponding segmental multifidi (130). The initial population of 41 patients accepted into the study and clinically felt to have facet-mediated pain was reduced to only 15 candidates for actual RFA, which further emphasizes the importance of double-block screening for Z-joint pain diagnosis. A recent randomized study using single Z-joint blocks for selecting patients with subsequent radiofrequency neurotomy demonstrated modest success with an average VAS pain reduction of 2.0 points in 66.7% of the lesion group versus 35.7% of sham group patients at 8 weeks to 12 months follow-up (131). An obvious concern with this study is that the criteria for a positive block mandated a relatively low threshold of only 50% pain relief, and 40 of 92 patients subjected to the diagnostic block had a positive response in this regard. This would imply a prevalence of facet joint pain higher than what has been reported in most populations. Another study showed no treatment effect at 12 weeks, as assessed by the Roland-Morris scale (2.6% change) and Oswestry scale (1.9% change) and VAS (–7.6% change) (132). Again, the known 38% false-positive rate for a single-block makes the study results and conclusions less convincing (126,133).
Efficacy of radiofrequency neurotomy for cervical Z-joint pain (other than from the C2-3 joint that was excluded from this study), has also been demonstrated in a randomized, double-blinded, and placebo-controlled trial (134,135). The patients were selected by placebo-controlled medial branch blocks. The total duration of pain relief was defined as the period until the patient judged that pain had returned to 50% of the pre-procedural level. Twenty-four patients were randomized into radiofrequency neurotomy and sham radiofrequency neurotomy (radiofrequency probes placed but radiofrequency was not turned on) groups. Fifty percent pain relief lasted 263 days in the radiofrequency group and 8 days in a controlled group. A second study of 28 patients with long-term follow-up and repeat radiofrequency neurotomy demonstrated a median duration of pain relief of 422 days. In the 11 patients who underwent repeat radiofrequency neurotomy, the median duration of pain relief was 219 days. The authors concluded that radiofrequency neurotomy provides clinically significant pain relief, and can be repeated if pain recurs (136).
Recent review articles on randomized controlled trials of radiofrequency neurotomy for spinal pain concluded that RF neurotomy was efficacious for both chronic low back pain and neck pain after flexion-extension injuries (124), and that there is evidence of moderate strength for use of RFA in the cervical and lumbar spine (129).
Potential complications of fluoroscopic-guided, contrastenhanced lumbar Z-joint injections or medial branch blocks are rare. The most common post-procedural problem is transient pain at the injection site. However, there have been rare case reports of meningitis, inadvertent spinal anesthesia, and infection after Z-joint injections (137,138). Recurrent back and neck pain after radiofrequency neurotomy may be due to incomplete ablation or medial branch regeneration (139). Since the dorsal root ganglia are left intact, the ablated medial branches may regenerate. Because radio frequency neurotomy does not permanently denervate the Z-joints or cause inherent instability in the spine, the concern of Charcot facet joint development has little ground and no such cases have been published (140). Also, because the DRG remains intact, deafferentation pain should not occur. Side effects from radiofrequency neurotomy are rare when the procedure is performed correctly but include local pain and infection.