Cooled Versus Conventional Thermal Radiofrequency Neurotomy for the Treatment of Lumbar Facet–Mediated Pain

A 59-year-old woman with no significant past medical history presents to the spine clinic with right-sided lower back pain. She reports having pain off and on for 10 years with no history of trauma or injury. The pain is typically worse when she is standing in place, with some improvement upon walking and resolution of pain while sitting. The pain is localized to the right L3-L5 paraspinal and gluteal regions. Upon examination, her sensory, motor, and muscle stretch reflexes are all within normal limits. Pain is exacerbated by lumbar extension and quadrant loading toward the right. Magnetic resonance imaging of the lumbar spine reveals mild disk degeneration but evidence of facet arthropathy of the right L4/L5 and L5/S1 facet joints. She is diagnosed with lumbar facet–mediated pain and is scheduled for diagnostic medial branch blocks. After positive responses (>80% reduction in pain) to 2 sets of medial branch blocks, she presents to discuss the next steps. She recently saw an advertisement for cooled radiofrequency neurotomy (RFN) and would like to consider this option. Drs Prempreet Bajaj and Jonathan Napolitano will advocate for cooled RFN, arguing that this procedure results in improved patient outcomes and is a safe procedure. Drs Wenbao Wang and Jianguo Cheng will argue that evidence of the efficacy of cooled RFN is limited and that the increased costs do not justify its use. The case presented is a very common scenario in pain practices throughout the United States. Patients often present with chronic axial low back pain, likely originating from degenerative changes of the lumbar facet joints. The lumbar facet joint pain referral patterns most commonly affect the lower back and buttocks, less commonly affect the posterior thigh, lateral hip, and groin, and rarely affect the anterior thigh and lower lateral calf 1-3. Diagnosis is based on clinical history, physical examination findings (pain with extension and facet loading), and imaging evidence of degenerative changes of the facet joints. However, the most important and reliable tool used to diagnose facet-mediated pain is the medial nerve branch block. The diagnosis is confirmed with 2 separate diagnostic blocks with pain relief greater than 80% 4, 5. Typically, a patient with degenerative facet arthropathy confirmed by 2 diagnostic blocks would be a good candidate for radiofrequency neurotomy (RFN) 4-6, assuming other treatment modalities, including physical therapy and anti-inflammatory medications, have not been successful. Traditional RFN has been used by pain physicians to mitigate pain stemming from the facet joints for many years. However, to understand the limitations of traditional RFN, we must understand the anatomy of the facet joint and its innervations. Recent studies have demonstrated that the medial branches of the lumbar facet joints (L1-L4) commonly lie across the lateral aspect of the superior articular process (SAP) at its neck, with some lying within the groove at the junction of the SAP and transverse process (TP). This anatomy is more variable at the L5 dorsal ramus 6, 7. It is also important to understand the location and orientation of the mamillo-accessory ligament (MAL). Distal to the neck of the SAP, the medial branch disappears under the MAL. Therefore, placing a lesion across the posterior quarter of the neck of the SAP is superfluous, because the MAL will likely protect the nerve from neurotomy. The MAL may also ossify, converting into a calcified bony element, making it even more difficult to lay a traditional RFN probe parallel to the nerve before it dives under the MAL 8. Furthermore, to obtain a complete denervation, typically a 2-burn lesion is necessary to capture these anatomic variances—one along the inner groove of the SAP to cover the nerves located slightly rostral on the SAP, and a second along the outer groove to cover the nerves located more laterally along the SAP/TP. Using a 2-burn lesion technique has become standard practice. For traditional RFN, our practice uses a temperature of 80°C for a burn time of 105 seconds (ramp time of 15 seconds and lesion time of 90 seconds). After repositioning the probe, the total time of a 2-burn lesion is 210 seconds. Anatomically speaking, the typical patient with chronic low back pain from facet arthropathy would have significant degeneration of the spine, with osteophytes and calcifications altering the anatomy and disrupting access for the RFN probe along the medial branch nerve. Thus, more fluoroscopic time is often added to obtain optimal images to visualize needle placement. Although the probe is in place, it may not be perfectly parallel to the nerve as a result of the osteophytes, limiting the efficacy of the lesion. This problem is especially seen in patients with significant scoliosis and degenerative changes in the spine. Understanding the limitations of traditional RFN explains why cooled RFN is a useful and more attractive tool that is now available as a treatment modality in pain management. Cooled RFN is a technology initially developed for various applications in cardiology and local treatment of solid malignancies. The physics behind cooled RFN versus traditional RFN allows for a much larger burn area, typically 2 times in diameter and by a factor of 8 in spherical circumference (because the burn area generated is a sphere) 9. Thus, using an 18-gauge 4-mm traditional RFN needle, the burn area diameter is 5 mm, whereas with a cooled RFN needle of the same size, the burn area diameter is 10 mm. Also, whereas a traditional RFN lesion is elliptical in shape, requiring the probe be precisely placed parallel to the nerve, the cooled RFN needle can be placed perpendicular to the nerve (similar to the approach of a medial nerve branch block) because the burn area is a sphere and the needle does not need to lie directly on the nerve. Technically this approach is easier, but it is also much more efficient in terms of fluoroscopic and procedure times. The technique introduces the feasibility of treatment for patients with severe degenerative changes of the spine and facet joints, where it is difficult to lay the traditional RFN probe. In those patients—a large percentage of the patients who have chronic back pain—the cooled RFN probe can easily be placed perpendicular to the nerve with the needle tip aligned along the mid portion of the SAP, just medial to the SAP/TP border. This positioning allows the burn lesion to capture the anatomic variants of the medial branch both superior and lateral to the SAP. Another point to consider is the patient's ability to tolerate the RFN procedure. With traditional RFN (in our practice), the temperature is 80°C and the time it takes to create the lesion is 210 seconds for each medial branch. Comparatively, for cooled RFN, the temperature is 60°C with a total lesion time of 150 seconds (25 seconds ramp time and 125 seconds burn time) for each nerve. This difference is compounded further for multilevel lesions. For example, in the case presented, the patient had facet arthropathy of the L4-L5 and L5-S1 facet joints, calling for RFN of the L3 and L4 medial branches and the L5 dorsal ramus. Thus 3 lesions are required for a unilateral treatment of this patient. In traditional RFN, a total burn time of 630 seconds would be required, whereas for cooled RFN, the total burn time would be 450 seconds. One also must consider how much the C-arm needs to be manipulated during the procedure. With traditional RFN, to achieve the perfect parallel alignment of the probe along the nerve, imaging begins with a caudal tilt followed by alignment of the end plates for each vertebral body to obtain best placement of the probe. This process can add undesired fluoroscopic time. Alternately, in cooled RFN, after aligning the end plates, an oblique view of the SAP allows placement of the probe in a gun barrel technique along the inferior and medial border of the SAP. We recently completed a small retrospective study of 60 patients in our clinic (abstract accepted for presentation at the AAPM&R Annual Assembly in October 2015 10) comparing fluoroscopic times of these procedures and found that the fluoroscopic time of traditional RFN was significantly longer than that of cooled RFN. This discrepancy was most evident when 6 lesions were required: 196.2 seconds with traditional RFN versus 141.4 seconds with cooled RFN. Given the increase in C-arm manipulation, increased burn time, and increased fluoroscopic time, traditional RFN takes a significantly longer time to complete from patient entry to recovery compared with cooled RFN. Cost must be considered, but when we examine the benefits that cooled RFN provides compared with traditional RFN, the costs of purchasing the cooled RFN equipment is money well spent, with greater efficiency and possibly better outcomes, although prospective studies directly comparing the 2 modalities have yet to be performed. In addition to using cooled RFN for the lumbar facet joints, once purchased, the equipment can also be used in multiple other areas where traditional RFN may not be used, such as the sacroiliac (SI) joint, knee, and hip, allowing the pain provider to be more valuable to his or her referral sources. This case represents a scenario that is commonly encountered in pain management and PM&R spine clinics. The patient has chronic right-sided axial low back pain. Magnetic resonance imaging reveals evidence of facet arthropathy of the right L4-L5 and L5-S1 facet joints. The diagnosis was further confirmed by 2 diagnostic medial branch blocks with greater than 80% pain reduction. Conventional thermal RFN of the medial branches is the standard of care. We argue against using cooled RFN for this case. The zygapophysial or “facet” joints are a set of synovial plane joints between the articular processes of 2 adjacent vertebrae. Lumbar facet joints are a well-recognized source of low back and referred pain to the lower extremity in patients with chronic low back pain 1. The facet joint is innervated by medial branches of the dorsal rami at L1-L4 levels and the dorsal ramus at the L5 level. Each lumbar medial branch innervates 2 adjacent facet joints. RFN involves the use of electrical current to generate a controlled lesion through which to safely interrupt nociceptive input. Numerous descriptive studies and 6 randomized clinical trials have reported the clinical efficacy and safety of lumbar RFN 2-7. At least 2 systematic reviews concluded that there is moderate evidence that radiofrequency denervation provides better pain relief than sham intervention 8, 9. Conventional RFN is the standard of care for facetogenic pain and can provide from 6 months to, occasionally, more than 12 months of relief. Conventional RFN is recommended by many professional society guidelines. After a comprehensive review of the literature, the American Society of Interventional Pain Physicians published its guidelines in 2013 and stated, “There is good evidence for conventional radiofrequency neurotomy for the treatment of chronic lumbar facet joint pain both in short-term and long-term” 1. The American Association of Neurological Surgeons and the Congress of Neurological Surgeons Joint Committee published their guidelines in 2014 and stated, “There is moderate evidence that facet medial nerve neurotomy (conventional RFN) produces a short-term decrease of facet-mediated chronic low-back pain (level II evidence)” 10. Cooled RFN was first introduced to pain medicine to treat pain arising from the SI joint 11, 12. It has also been suggested for treatment of thoracic facet joint–mediated pain 13. Using the water cooling system, cooled RFN is thought to create a bigger lesion than conventional RFN. A larger lesion may be necessary to overcome the anatomic variations of the lateral branches of the sacral nerves innervating the SI joints and the medial branches innervating the thoracic facet joints. Two randomized controlled trials and a few case series have been reported for its use in treating SI joint pain and demonstrated efficacy. However, a comparative study did not reveal a significant advantage of cooled RFN compared with conventional thermal RFN 14. Nevertheless, some clinicians believe it may be used to treat facetogenic pain and achieve better results. Use of cooled RFN is not necessary for 2 reasons. First, the anatomy of the medial branches in the lumbar spine is relatively consistent. Human cadaveric studies have revealed that the medial branches of the dorsal rami at segmental levels L1-L4 assume a constant and similar course 15-18. Each nerve emerges from its intervertebral foramen and enters the posterior compartment of the spine by coursing around the neck of the superior articular process below the foramen. The medial branch passes caudally and slightly dorsally to disappear under the MAL, where the nerve hooks medially around the caudal aspect of the root of the superior articular process to enter the multifidus muscle, in addition to innervating the facet joint 18, 19. The consistent course of each medial branch is due to 2 anatomic features. Its point of entry into the posterior compartment is fixed through a foramen in the posterior leaf of the intertransverse ligament immediately rostral to the junction of the SAP and TP. Caudally, the nerve is again fixed by the MAL, which allows for virtually no variation in the location or orientation of the nerve as it crosses the neck of the SAP 18, 19. Based on these anatomic findings, if the RFN electrodes are placed accurately at the junction of the superior articular process and transverse process, the result should be predictable. Unfortunately, for historical reasons, flawed techniques have been widely performed even in frequently cited randomized controlled trials 5. Shealy 20 initially described the articular branches of the medial branch nerve to the lumbar zygapophysial joints as rising dorsally across the lateral aspect of the superior articular processes. He recommended placing electrodes lateral to the articular process, perpendicular to the coronal plane 20. However, studies by Bogduk et al have shown that the medial branch of each lumbar dorsal ramus runs caudally across the neck of the superior articular process and that articular branches approach each joint from below and from above 15, 19, 21. There was no nerve rising dorsally through the region where Shealy recommended placing electrodes 15, 18, 19. The study by Lau et al 18 shows that when electrodes are placed “end-on” against the nerve, most of the thermal lesion lies away from the nerve. Even if the tip of the electrode lies near, or even on, the nerve, because of the tapering shape of the lesion, the nerve may be spared from coagulation or be only partially coagulated. Such placements may lead to no relief, partial relief, or only short-lived relief 18. The International Spine Intervention Society prescribed certain standards of practice for lumbar medial branch RFN. It recommended that, for lumbar medial branch neurotomy (MBN) to be anatomically accurate, electrodes should be placed parallel to the target nerve. The idea is not to make a larger lesion but to make a more precise one. Second, the major limitation of maximum radiofrequency (RF) power output is disruption of conductivity at the electrode-tissue interface 22, 23. When tissue at the electrode-tissue interface desiccates, boils, or chars, it becomes a high impedance insulator. This phenomenon underlies most of the or increase in impedance that can during RFN and will result in an The temperature at which these is as the interface disruption The of RFN is to the electrode tissue interface below the while power is to the lumbar the electrode-tissue interface temperature is controlled at 80°C to which is not even to and cooling is not Cooled electrodes are only when the is the limiting factor on The safety of using cooled RFN for lumbar has been our not a study of the use of cooled RFN for lumbar has been of this of literature, we in which cooled RFN was used for other pain One case of a burn from cooled RFN of the medial branch nerve for treatment of thoracic facet was reported et al in the during cooled RFN for vertebral lesions. The temperature increase to when a tip was used, which is high to In the lateral and branches of the dorsal which the and are not away of those nerves may result in muscle which may of the lumbar spine and the back pain. We argue against using cooled RFN for lumbar facet we using conventional RFN to achieve controlled lesions of the medial branch cooled RFN is an procedure. The cooled RFN costs more than Lumbar is the second most commonly performed procedure in the United and the use of cooled RFN will increase the by of of care for an procedure. In conventional RFN for lumbar medial branches is the standard of care for lumbar facetogenic pain. It is safe and evidence to the and of cooled RFN for lumbar Even in no advantage in using cooled RFN than conventional RFN. At this cooled RFN for lumbar should be Patients should be of this when are in such Drs Wang and Cheng that a comparative study by Cheng and no advantage of cooled compared with traditional of the SI study had many First, the study outcomes were only by pain and not by outcomes or the use of pain significant between the 2 in to the patient and spine significant also between the with to treatment intervention of lesions of levels and Furthermore, treatment for cooled did not the current recommended in of the was no of which nerves were and which procedures were performed. because of the retrospective of the the not many including how many performed the procedure and how patients compared with In a large systematic review of SI joint that the evidence was for traditional radiofrequency but for cooled of the SI results were directly to the larger burn area that the cooled probe generated to for the in the SI joint The study by Cheng and a good in comparing the 2 but his based on the evidence were not With all the it was like comparing to Drs Wang and Cheng also that cooled is not necessary for lumbar because of the of the nerve A study performed by and 3 with cadaveric demonstrated that variation in the location of the medial and lateral branches of the dorsal ramus. This variation was most in the lumbar spine but was also in the lower lumbar including most significantly at the facet joint we stated, placing a probe parallel to the medial branch nerve may be difficult because the MAL can This seen in of the can lead to in aligning the probe for placement 4, 5. Drs Wang and Cheng that studies performed for had limitations because of the techniques used by pain pain have precise probe or those who are just this procedure to their With use of the cooled system, the lesion and location of the probe allows for a large burn Cooled as a treatment for low back pain is as safe as traditional RFN. Drs Wang and Cheng have also stated, there are no or studies that cooled is less safe than traditional in the treatment of low back pain. Cooled can also be used for a of pain management procedures including treatment of the SI joint, hip, and pain. With better results that are directly to the large burn area, it is difficult to cooled on the of It is with that we the by Drs Bajaj and Napolitano on this Unfortunately, we have to that we with on First, that studies have demonstrated a variable course of the at However, the of their are more than 10 years and only of the this For example, cited the by Lau et al as evidence of this we cited the same during our In as in our the of the nerve by the mamillo-accessory ligament allows for virtually no variation in the location or orientation of the nerve as it crosses the neck of the superior articular in the we that the anatomy is significantly at the L5 level. Second, the medial branches at the lumbar levels are for only a limited for safe and of lateral to the of the neck of the superior articular process 1. Distal to this area, the nerve lies under the MAL. to this area, the medial branch lies to the and of the and lateral branches of the dorsal ramus. When using a tip needle, lesions in 2 are However, with an electrode with a a lesion this is typically of the MAL is because the of the nerve before the MAL is by RFN. With imaging and it is typically to RFN. In it is of the most commonly performed procedures across the United with an safety In the efficacy and safety of cooled have not been It is of to use cooled to create a large lesion in the area, which is with and the by Drs Bajaj and Napolitano that high temperature may electrode interface disruption is The common practice is to use which is below the of The Drs Bajaj and Napolitano cited is for The was 12 at maximum tissue temperature of which is to the Thus cooling is not necessary for lumbar medial branch RFN. Drs Bajaj and Napolitano that cooled RFN is a procedure. However, their is In traditional RFN is typically performed in much less time than cooled RFN. With multiple probe to electrodes can be only 105 seconds to complete the lesion, in to a of 450 seconds for cooled RFN. Even with 2 of traditional RFN takes less time than does cooled RFN. but not Drs Bajaj and Napolitano argue that the of purchasing the cooled RFN is money However, to a very important The total is not only the of the cooled RFN needle a of or more for case. It is not to use cooled RFN with the current of for this procedure. We believe it is not to a practice based only on We are to only if it is by solid on and which is cooled RFN for lumbar facet denervation is to low results are not

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.