Abstract

Objective. To determine the effectiveness and risks of fluoroscopically-guided cervical medial branch thermal radiofrequency neurotomy (CMBTRFN) for treating chronic neck pain of zygapophysial joint origin.

Design. Systematic review of the literature with comprehensive analysis of the published data.

Interventions. Four reviewers formally trained in evidence-based medicine searched the literature on CMBTRFN. Each assessed the methodologies of studies found and appraised the quality of evidence presented.

Outcome Measures. The primary outcomes assessed were 100% relief of pain 6 and 12 months after treatment. Other outcomes were noted if reported. The evidence was evaluated in accordance with the Grades of Recommendation, Assessment, Development, and Evaluation (GRADE) system.

Results. The searches yielded eight primary publications on the effectiveness of the procedure. The evidence shows a majority of patients were pain free at 6 months and over a third were pain free at 1 year. The number needed to treat for complete pain relief at 6 months is 2. The evidence of effectiveness is of high quality according to the GRADE system. Twelve papers were found reporting unwanted effects, most of which are minor and temporary. No serious complications have ever been reported from procedures performed according to the published guidelines. The evidence of risks is of low quality according to the GRADE system.

Conclusions. If performed as described in the International Spine Intervention Society Guidelines, fluoroscopically-guided CMBTRFN is effective for abolishing zygapophysial joint pain and carries only minor risks.

Introduction

Cervical medial branch thermal radiofrequency neurotomy (CMBTRFN) is a minimally invasive percutaneous procedure for the treatment of chronic neck pain and cervicogenic headache. It uses an electric current, alternating at the frequency of radio waves, to coagulate the nerves from cervical zygapophysial (Z) joints, and thereby interrupt nociceptive transmission from these joints. The target nerves are the medial branches of one more of the cervical dorsal rami, or the third occipital nerve, which have previously been shown to the responsible for mediating the patient's pain. A specialized electrode is placed immediately adjacent to, and parallel to, the target nerve. A radiofrequency generator delivers the alternating current to the body through a dispersion pad. The resulting electric field concentrates near the tip of the electrode, and heats the tissues surrounding the tip, thereby coagulating the adjacent nerve. The process is repeated to produce multiple radiofrequency (RF) lesions along each target nerve. The nerves are treated one at a time, and a maximum of three nerves treated in any one application of the procedure.

Critical to correct interpretation of the evidence on CMBTRFN is understanding that there are various other procedures with similar names, which are often mistakenly lumped together. For the purposes of this review, CMBTRFN is defined as the procedure described in the practice guidelines published by the International Spine Intervention Society in book form [ 1 ] and in antecedent journal articles [ 2 , 3 ].

The history of RF treatments goes back more than 50 years. Radiofrequency neurotomy (RFN) was first described as a treatment for pain in 1960, in a paper on relief of intractable pain by percutaneous anterolateral cordotomy [ 4 ]. RFN was subsequently applied as a treatment for trigeminal neuralgia in 1967 [ 5 ]. In 1970s and 1980s, RFN began to be used to treat back pain and neck pain.

For the treatment of back pain Shealy [ 6 ] and Schaerer [ 7 , 8 ] described thermal RF techniques aimed at the center of the target joint. Sluijter [ 9 ] wrote about techniques aimed at the dorsal ramus and the dorsal root ganglion. Bogduk and Long [ 10 ] demonstrated that the techniques of Shealy and Schaerer were not anatomically valid, and corrected the technique to target the lumbar medial branches accurately [ 11 ].

For the treatment of neck pain, a review of the literature [ 12 ] highlighted the limitations of earlier studies and outlined a sound anatomical basis for cervical medial branch neurotomy.

Subsequent studies validated the procedure [3,4,13,14], and became the basis for the guidelines of the International Spine Intervention Society (the Society) [ 1 ].

However, other forms of RF treatment, many of them not anatomically valid, remain in use by some; and those procedures are often confused with CMBTRFN as defined by the Society. These other procedures include those that purport to place effective thermal RF lesions over target nerves but are performed in ways that do not achieve the purpose, and procedures with similar-sounding names such as “pulsed RF neurotomy” in which radiofrequency currents are used to produce heat of much less intensity from electrodes placed quite differently. Such procedures are outside the scope of this review.

The Society's description of CMBTRFN includes the indications for the procedure [1]. Specifically, it should be done only for Z joint pain diagnosed definitively by analgesic responses to comparative medial branch blocks (MBBs) or third occipital nerve blocks (TONBs). The validity of these tests has been demonstrated in three domains: face validity, construct validity, and predictive validity. Their face validity, or target-specificity, was demonstrated in a study showing the spread of the injectate is limited to the location of the target nerve [ 15 ]. The construct validity of MBBs depends on the extent to which positive or negative block results show that the joint tested is, or is not, a pain source. Single cervical MBBs have a false-positive rate of 27% [ 16 ]. This liability of false-positive results is overcome by undertaking comparative MBBs, i.e., performing two blocks on separate occasions using different local anesthetics and comparing the durations of response to the durations of action of the agents injected [ 17 ]. The addition of a placebo control provides even greater diagnostic confidence, but the utility of comparative cervical MBBs has been shown to be satisfactory for clinical purposes [ 18 ]. Thus, in this review, “comparative blocks” include dual blocks performed with or without the addition of placebo controls. Predictive validity refers to the value of diagnostic test results in predicting responses to treatment. The predictive validity of cervical MBBs was established in a study which showed positive and concordant responses to comparative MBBs accurately predicted successful outcomes when those joints were treated by RFN [ 2 ]. Thus, the diagnostic validity of comparative cervical MBBs has been demonstrated in all three domains.

Other methods of patient selection for RF treatment have been described, including tenderness to palpation on clinical examination [ 19 , 20 ] and analgesic response to joint blockade by intra-articular injection of local anesthetic [ 11 ] or dorsal ramus blockade [ 21 ]. None of these methods of patient selection have been validated by sound evidence, and in particular, clinical examination has been shown to be invalid [ 22 ].

Medical imaging modalities have been investigated as a means of diagnosing cervical joint pain, but none has been validated. Radiographic appearances simply do not correlate with cervical pain. Studies have shown that plain X-ray findings [ 23–27 ], computerized tomography (CT) findings [ 23 , 26 ], even when single photon emission CT is used [ 28 ], and the findings of magnetic resonance imaging [ 26 , 29–33 ] are not useful for diagnosing painful cervical joints. No imaging modality provides valid diagnosis of a painful cervical joint.

Accordingly, positive results of comparative MBBs or TONBs constitute the only validated methods for diagnosis of cervical Z joint pain and, therefore, for selecting patients for CMBTRFN [ 17 , 34 ]. Procedures done without those indications are outside the scope of this review.

The importance of the effectiveness of CMBTRFN is underlined by the prevalence of cervical Z joint pain. Studies have shown cervical Z joints are the most common sources of pain associated with motor vehicle accidents which cause “whiplash” injuries [ 35–37 ]. Studies using comparative MBBs and TONBs show cervical Z joint pain accounts for 60% (CI 95 46–73%) of chronic neck pain after whiplash [ 37 ], and when headache is the dominant symptom after whiplash, pain referred from the C2-3 Z joint (third occipital headache) accounts for 53% (CI 95 37–68%) of cases [ 35 ]. Cervical Z joint pain also has a high prevalence in nontraumatic cases of chronic neck pain [ 38 ]. These data suggest cervical Z joint impairment is a common cause of chronic neck pain and cervicogenic headache; so, CMBTRFN is a treatment that is commonly indicated. This review identifies and appraises the published evidence on its effectiveness and its risks.

Methods

Four investigators, who all have formal training in evidence-based medicine, searched the scientific literature independently for publications on the effectiveness of fluoroscopically-guided CMBTRFN. Initially they each conducted digital searches using the search engine Ovid to explore the databases Embase, Medline, and EBM Reviews, using the keywords cervical, zygapophysial, facet, medial, branch, radiofrequency, and neurotomy. The searches encompassed all scientific papers published until March 2015 but excluded were non-human studies, conference abstracts and case reports, unless they were reports of complications. When suitable papers were retrieved, the references of each were perused for relevant citations that might not have been identified by the database searches.

The primary publications found were classified into three types: observational studies (simply describing outcomes after the use of an intervention), pragmatic studies (comparing outcomes after use of an intervention with outcomes after use of another intervention expected to have a therapeutic effect), and explanatory studies (comparing outcomes after use of an intervention with outcomes after use of another intervention not expected to have a therapeutic effect). The primary papers on effectiveness of fluoroscopically-guided CMBTRFN were appraised by each of the investigators independently, using an instrument developed by the International Spine Intervention Society's Standards Division to facilitate reliable assessment of studies of therapeutic effectiveness. The investigators then discussed the studies to reach consensus on the value of each paper's contribution to the published evidence of the effectiveness of CMBTRFN.

Categorical data were sought as the preferred evidence of effectiveness because they reflect binary outcomes (patients achieving or not achieving a successful result). Such data, expressed as success rates, can be collated to produce a body of evidence of effectiveness based on outcomes for specific patients [ 39 ]. In this review, the primary outcome measures sought were success rates for complete relief of pain at 6 and 12 months after treatment. The results of studies that produced categorical data for individual patients were tabulated.

The data from all the studies following the Society's guidelines were appraised and the resultant body of evidence was evaluated using the Grades of Recommendation, Assessment, Development, and Evaluation (GRADE) system of appraisal to determine the quality of the evidence of the effectiveness of CMBTRFN.

Using the same methodology, the published data on the risks of CMBTRFN were evaluated using the GRADE system. The same keywords were used, and safety, complications, and unwanted effects were added. The quality of the published evidence of effectiveness and the published evidence of risks were then both taken into account and conclusions were drawn in accordance with the GRADE system about the strength of recommendations for use of CMBTRFN based on all published data on the procedure.

Results

The literature searches yielded eight primary papers that reported data on effectiveness. Of the eight reports, six were observational studies and two were explanatory studies. Also found were 12 papers reporting complications of cervical RFN, but only four of them related to CMBTRFN as defined. Another 76 publications were identified but were not included in this review because the patients were selected by invalid methods, the papers contained no primary data, or the papers were letters or reviews of the literature.

Effectiveness

Observational Studies

The first data on CMBTRFN using dual comparative blocks with complete relief for patient selection was an observational pilot study [ 12 ]. Nineteen patients and 21 joints were treated. The data were divided into outcomes of RF treatment of the third occipital nerve (TON) for chronic pain from the C2-3 Z joint (n = 10 patients) and treatment of the C5, C6, and C7 medial branches for chronic pain from the C5-6 and C6-7 Z joints (n = 10 patients); one patient had the C2-3, C5-6, and C6-7 joints treated. As the third occipital neurotomies were not done in accordance with the Society's current guidelines, their results were not suitable for inclusion in this review. After CMBTRFN for lower cervical Z joint pain, 7/10 patients (70% [CI 95 42–98%]) were pain free at 6 months and 4/10 patients (40% [CI 95 10–70%]) were pain free at 1 year. The authors of this pilot study concluded that the results justified progress to a randomized controlled trial of CMBTRFN for lower cervical Z joint pain.

Five more observational studies of the effectiveness of CMBTRFN were published between 1998 and 2012 [ 3 , 13 , 14 , 40 , 41 ]. In all of these studies, patients were selected based on complete relief of chronic neck pain and/or headache by each of two CMBBs or TONBs, and were treated by CMBTRFN performed strictly in accordance with the Society's guidelines. All patients were followed up assiduously, and their outcomes were recorded to determine the success rates of the treatment.

The first and second of these studies were from the same group of investigators and addressed different aspects of CMBTRFN when performed on the same group of 28 patients with chronic neck pain stemming from Z joints between the C3-4 and C6-7 levels [ 13 , 14 ]. At review 6 months after treatment, 16/28 (57% [CI 95 39–75%]) were pain free. After 12 months, 10/28 (36% [CI 95 18–54%]) were pain free. The median duration of relief after a first procedure was 219 days when failures were included; but for patients who had a successful outcome, the median duration of relief was 422 days. An important finding was that when the effect wore off and the pain recurred, pain relief could be reinstated by repeat CMBTRFN. Another important finding was that the litigation status of patients did not influence outcomes.

A study published in 2003 addressed the treatment of chronic upper neck pain and cervicogenic headache stemming from C2-3 Z joints [ 3 ]. This study described a new TON treatment procedure, which is now the basis for the Society's guidelines, and was designed to overcome the technical problems identified in an earlier pilot study [ 12 ]. Of the 49 patients included, 32 (65% [CI 95 52–78%]) were pain free at the 6 month review and 10 (20% [CI 95 9–31%]) were pain free after 12 months.

Two studies published in 2005 and 2012 investigated the effectiveness of CMBTRFN when applied in community settings to treat chronic pain from any of the cervical Z joints. In the earlier study involving 35 patients, 16 (46% [CI 95 29–63%]) were pain free at 6 months and nine (26% [CI 95 11–41%]) were pain free at 12 months [ 40 ]. In the later study involving 104 patients, 71 (68% [CI 95 59–77%]) were pain free at 6 months and 53 (51% [CI 95 41–61%]) were pain free at 12 months [ 41 ].

Explanatory Studies

In 1996, Lord et al. published a double blind explanatory study of the effectiveness of CMBTRFN performed on patients with chronic cervical Z joint pain diagnosed by comparative, placebo-controlled MBBs [ 2 ]. At 6 months after CMBTRFN, 7/12 (58% [CI 95 30–86%]) patients in the active treatment group were pain free, while 1/12 (8% [CI 95 0–23%]) patients in the placebo group was pain-free. These confidence intervals do not overlap, demonstrating that the results of CMBTRFN are not due to a placebo effect. These data also show the number needed to treat (NNT) by CMBTRFN for complete pain relief is 2.

The following year, the same group of investigators published another explanatory study of the effect of CMBTRFN on psychological distress [ 42 ]. The results were measured only 3 months after treatment, so are not suitable for inclusion with the other data in this review. Patients in the active treatment group who experienced complete relief of pain also had resolution of associated psychological symptoms.

The categorical data on the effectiveness of CMBTRFN after 6 months and 12 months are tabulated below in Tables 1 and 2 .

Table 1

Reported rates of success (defined as abolition of pain) 6 months after CMBTRFN of patients selected by analgesic responses to comparative medial branch blocks

Study Design Levels Pain Completely Relieved 
Lord [1995] Observational C5-7 7/10  70% (CI 95 42–98%)  
Lord [1996] Explanatory C3-7 7/12  58% (CI 95 30–86%)  
McDonald [1999] Observational C3-7 16/28  57% (CI 95 39–75%)  
Govind [2003] Observational C2-3 32/49  65% (CI 95 52–78%)  
Barnsley [2005] Observational C2-7 16/35  46% (CI 95 29–63%)  
MacVicar [2012] Observational C2-7 71/104  68% (CI 95 59–77%)  
Study Design Levels Pain Completely Relieved 
Lord [1995] Observational C5-7 7/10  70% (CI 95 42–98%)  
Lord [1996] Explanatory C3-7 7/12  58% (CI 95 30–86%)  
McDonald [1999] Observational C3-7 16/28  57% (CI 95 39–75%)  
Govind [2003] Observational C2-3 32/49  65% (CI 95 52–78%)  
Barnsley [2005] Observational C2-7 16/35  46% (CI 95 29–63%)  
MacVicar [2012] Observational C2-7 71/104  68% (CI 95 59–77%)  
Table 2

Reported rates of success (defined as abolition of pain) 12 months after CMBTRFN of patients selected by analgesic responses to comparative medial branch blocks

Study Design Levels Pain Completely Relieved 
Lord [1995] Observational C5-7 4/10  40% (CI 95 10–70%)  
McDonald [1999] Observational C3-7 10/28  36% (CI 95 18–54%)  
Govind [2003] Observational C2-3 10/49  20% (CI 95 9–31%)  
Barnsley [2005] Observational C2-7 9/35  26% (CI 95 11–41%)  
MacVicar [2012] Observational C2-7 53/104  51% (CI 95 41–61%)  
Study Design Levels Pain Completely Relieved 
Lord [1995] Observational C5-7 4/10  40% (CI 95 10–70%)  
McDonald [1999] Observational C3-7 10/28  36% (CI 95 18–54%)  
Govind [2003] Observational C2-3 10/49  20% (CI 95 9–31%)  
Barnsley [2005] Observational C2-7 9/35  26% (CI 95 11–41%)  
MacVicar [2012] Observational C2-7 53/104  51% (CI 95 41–61%)  

If the data in each of Tables 1 and 2 are added, they show that when patients were treated by CMBTRFN after diagnosis by at least dual comparative blocks with 100% relief, 149/238 patients (63% [CI 95 57–69%]) were pain free at 6 months and 86/226 (38% [CI 95 32–44%]) were pain free at 1 year.

GRADE Evaluation of Evidence of Effectiveness

The two explanatory studies [ 2 , 42 ] start as high quality evidence and there is no reason to downgrade them. The six observational studies start as low quality evidence and there is no reason to downgrade any of them, but they should be upgraded for magnitude of effect and a dose response curve [ 43 ]. As the confidence intervals of the explanatory studies overlapped the confidence intervals of the observational studies, and data from the treated groups did not overlap the data from the placebo groups, the observational studies can be upgraded. In addition to the large magnitude of effect, there was a clear dose response relationship: when RF treatment effects wear off, complete pain relief can be reinstated with repeat treatment.

Thus, the evidence of the effectiveness of CMBTRFN for producing total relief of neck pain or cervicogenic headache (when patients are selected by CMBBs) is of high quality: the true effect is likely to be close to that shown by the published data, and further research is unlikely to change the conclusions presented.

Risks

The literature searches produced four papers that described risks associated with CMBTRFN as defined by the Society's criteria. Those risks are summarized in Table 3 .

Table 3

Risks and complications of CMBTRFN procedures recorded in the literature, their prevalence (95% confidence intervals), and the references in which they were first documented

Risks Levels Frequencies Notes References 
Radiation risks C2-7 100% Minimal Lord [1998] 
Postprocedural pain C2-7 98% (95–100%) Temporary Lord [1998] 
Cutaneous numbness C2-3 97% (93–100%) Enduring Govind [2003] 
C2-7 47% (36–58%) Lord [1998] 
Dizziness and ataxia C2-3 95% (90–100%) Temporary Govind [2003] 
C2-7 23% (14–32%) Lord [1998] 
Dysesthesias C2-3 55% (43–67%) Temporary Govind [2003] 
C2-7 30% (20–40%) Lord [1998] 
Pruritis C2-3 9% (2–16%) Temporary Govind [2003] 
Postprocedural infection C2-7 3% (0–9%) Temporary; one case Barnsley [2005] 
Vasovagal syncope C2-7 2% (0–5%) Temporary Lord [1998] 
Neuritis C2-7 2% (0–5%) Temporary to enduring Lord [1998] 
Dermoid cyst C2-7 1% (0–3%) Enduring: one case Lord [1998] 
Köbner's phenomenon C2-7 1% (0–3%) Enduring: one case Lord [1998] 
Risks Levels Frequencies Notes References 
Radiation risks C2-7 100% Minimal Lord [1998] 
Postprocedural pain C2-7 98% (95–100%) Temporary Lord [1998] 
Cutaneous numbness C2-3 97% (93–100%) Enduring Govind [2003] 
C2-7 47% (36–58%) Lord [1998] 
Dizziness and ataxia C2-3 95% (90–100%) Temporary Govind [2003] 
C2-7 23% (14–32%) Lord [1998] 
Dysesthesias C2-3 55% (43–67%) Temporary Govind [2003] 
C2-7 30% (20–40%) Lord [1998] 
Pruritis C2-3 9% (2–16%) Temporary Govind [2003] 
Postprocedural infection C2-7 3% (0–9%) Temporary; one case Barnsley [2005] 
Vasovagal syncope C2-7 2% (0–5%) Temporary Lord [1998] 
Neuritis C2-7 2% (0–5%) Temporary to enduring Lord [1998] 
Dermoid cyst C2-7 1% (0–3%) Enduring: one case Lord [1998] 
Köbner's phenomenon C2-7 1% (0–3%) Enduring: one case Lord [1998] 

The first study to mention risks of CMBTRFN was published in 1995 [ 12 ]. The authors did not provide specific data, but noted that after CMBTRFN procedures patients often experienced temporary dizziness, giddiness, and ataxia due to effects on nerve fibers involved in proprioception and balance. They also noted most patients needed analgesic medications for postoperative pain for some days after a CMBTRFN procedure. The paper also mentioned several theoretical risks of cervical RFN but they remain theoretical, as they have never been reported in the literature as having actually occurred.

Another study published in 1998 reported rates of unwanted events with CMBTRFN of the C3-7 Z joints [ 13 ]. It was an audit of 83 treatments completed over 5.5 years with meticulous follow-up of patients' outcomes. No serious complications occurred in the cohort. Less severe side-effects that were noted included postoperative pain in 81/83 or 98% (CI 95 95–100%), dysesthesias in 25/83 or 30% (CI 95 20–40%), cutaneous numbness in 39/83 or 47% (CI 95 36–58%), dizziness and ataxia in 19/83 or 23% (CI 95 14–32%), vasovagal syncope (or presyncope) and postprocedural neuritis were both reported in 2/83 or 2% (CI 95 0–5%), while development of a dermoid cyst and Köbner's phenomenon (psoriasis at the site of puncture) were each reported in 1/83 or 1% (CI 95 0–3%). Only one study has ever reported a postprocedural infection: Barnsley noted that, in one patient from a series of 47 CMBTRFN procedures performed on 35 patients, the infection was superficial and with oral antibiotics it resolved without sequelae [ 40 ].

Other theoretical risks not manifest in these early study cohorts but discussed in the papers as potential hazards included radiation exposure, allergic reactions, bruising, ground electrode burns, neuroma, anesthetica dolorosa, and risks associated with implanted electrical devices such as cardiac pacemakers.

Side-effects of CMBTRFN of the TON for C2-3 Z joint pain have also been reported in observational studies. Govind was the first to report rates of unwanted events in 2003 [ 3 ]. Numbness in a small area behind the ear (the area supplied by the TON) was reported afterward in 63/65 or 97% (CI 95 93–100%) of the procedures done. Temporary ataxia (lasting up to a day or so) was reported after 62/65 or 95% (CI 95 90–100%) of the procedures. Dysesthesias, ranging from hypoesthesia to hypersensitivity to touch, occurred after 36/65 or 55% (CI 95 43–67%) of procedures; the effects lasted for 1–2 weeks in most patients affected, but for 4 weeks in one case; no case required treatment. Itch occurred after 6/65 or 9% (CI 95 2–16%) of treatments; it too was of short duration.

The unwanted effects and risks of CMBTRFN reported in the literature are summarized below in Table 3 .

GRADE Evaluation of Evidence of Risks

When assessing the quality of the evidence on the risks of CMBTRFN in accordance with the GRADE system, it is noted that the published evidence consists of four observational studies. Accordingly, the body of evidence is of low quality: our confidence in the effect estimate is limited and the true effect may be substantially different from the estimate of the effect [ 43 ].

Discussion

Following the paradigm established by the Standards Division of the International Spine Intervention Society [ 44–47 ], this review is comprehensive: it includes the whole body of peer-reviewed literature producing categorical evidence and that body of evidence has been appraised using GRADE. If only randomized controlled trials had been studied, sound data from rigorous observational studies would have been ignored. This review does not include other techniques or selection criteria that have not been validated.

The evidence shows CMBTRFN results in 63% (CI 95 57–69%) of patients being pain free at 6 months and 38% (CI 95 32–44%) pain free at 1 year. It has also been shown to allow patients to return to all activities previously inhibited by pain, including work [ 48 ].

This effectiveness is absolutely dependent on the procedure being performed in accordance with the published guidelines [ 1 ], and in particular, on patients being selected following positive and concordant responses to dual, comparative, diagnostic blocks producing 100% relief (with or without additional placebo controls). The evidence for CMBTRFN cannot be generalized to RF procedures not performed in accordance with the guidelines; the effectiveness of such procedures is unknown.

The distinction between CMBTRFN performed according to guidelines and other types of RF treatment has not been recognized in previous reviews [ 49–51 ]. Those reviews included other types of RF treatment performed on patients selected in a variety of other ways, so it is not surprising that when results of such procedures are pooled with those of CMBTRFN performed in accordance with published guidelines, the overall picture is confusing and may be interpreted as showing RF treatments are ineffective. Some types of RF treatment undoubtedly are ineffective, but generalization is not justified. The reports of those previous reviews have given rise to misunderstanding of the effectiveness of CMBTRFN and have been used by some to disparage the procedure. The evidence of the current review counters that misunderstanding; it shows, unquestionably, that CMBTRFN is effective when performed in accordance with the Society's guidelines [ 1 ].

CMBTRFN is not a permanent cure for cervical Z joint pain. There is no intervention proven effective for restoring impaired spinal joints to their original condition. However, the actual impairments are usually very small and not of structural significance; their main effect is to cause chronic pain [ 24 , 25 , 52 ]. CMBTRFN abolishes the pain for long periods by denaturing the chemical components of nociceptive Aδ and C nerve fibers and preventing conduction through them. It does not destroy the target nerves. They remain intact anatomically and over time they recover; then the pain returns, although it can be abolished again by repeat treatment [ 3 , 13 , 40 , 41 , 53 , 54 ]. A major factor determining the duration of effect is the length of the target nerve coagulated through its full thickness in an RF treatment. The length of nerve coagulated is in turn influenced by variations in the neuroanatomy of the nerves, in particular, variations in their diameters and courses, and the sizes of lesions created in the procedure.

These factors are exemplified by the literature on third occipital headache. The initial results of thermal RFN for its treatment were disappointing [ 12 ]. The problem was that the TON is anatomically variable in both its size and its course. Anatomic studies showed the TON is generally larger than the other medial branch nerves, with a mean diameter of 1.5 ± 0.4 mm [ 55 ]. By comparison, the other medial branches have diameters ranging from 0.9 ± 0.3 mm (the C4 and C5) down to 0.5 ± 0.2 mm (the C3). In addition, the course of the TON has a wider range of anatomic locations than other medial branches [ 55 ]. These factors mean RF lesions placed over the TON have less chance of coagulating the full thickness of the nerve for a satisfactory length than do lesions at other levels. Knowledge of the anatomic variability led to development of a modified treatment technique, allowing for the variations by placing larger RF lesions over a wider range of positions. When that technique was used for thermal RFN of the TON, the treatment was shown to be as effective as CMBTRFN at other cervical levels [ 3 ].

The size of RF lesions is governed by the diameter (or gauge) of the RF electrode used to create them, the length of its active tip, the temperature it reaches, and the time the heat is applied by it. The effects of these factors on RF lesion size were addressed by Bogduk et al. in 1987 [ 56 ] and have been addressed again recently by Cosman et al. [ 57 ]. A 23 gauge SMK electrode with a 4 mm active tip heated to 80°C for 90 seconds produces an RF lesion averaging 3.9 mm wide and 5.9 mm long [ 13 ]; by comparison, a 16 gauge RRE electrode with a 6 mm active tip heated similarly creates an RF lesion averaging 9.2 mm wide and 10.5 mm long [ 57 ]. It is not hard to see why the 23 gauge SMK electrode used in the first TON study [ 12 ] was less effective than the 16 gauge RRE electrode used in the second [ 3 ]. The data show that a 16-gauge electrode is to be preferred for all CMBTRFN procedures.

The use of specific interventions for treatment of chronic pain is controversial in some quarters. Some practitioners apparently believe that when pain has been present for 3 months or more, central sensitization has occurred and that process is irreversible, so the pain will persist indefinitely. This belief is reflected in claims that chronic pain is a disease in its own right [ 58 , 59 ]. So-called “centralists” who hold that belief say attempts by what may be called “peripheralists” to treat nociceptive mechanisms of chronic pain are futile, and that chronic pain should be managed by “multidisciplinary” methods that are based largely on psychological techniques aimed at helping patients cope with the pain [ 60–62 ]. Such dichotomous views are unlikely to be consistent with the broad range of possibilities encompassed by medical science and are at odds with the biopsychosocial model of pain put forward by Engel [ 63 ] who suggested considering the psychological and social domains of pain in addition to, not instead of, its biomedical domain. The evidence of this review shows narrow “centralist” beliefs to be erroneous. When performed according to the Society's guidelines, CMBTRFN is an effective treatment for abolishing chronic pain, with an NNT of 2. It clearly reverses central sensitization and allows patients previously disabled by chronic neck pain or cervicogenic headache to resume life pain-free. The effect of RFN in countering central sensitization was demonstrated explicitly in a study published in 2014 [ 64 ]. Reversal of central sensitization is not unique to CMBTRFN: analogous abolition of chronic pain is a common outcome after hip or knee replacement surgery, and after many other treatments for painful conditions [ 65 ]. An explanatory study in 1997 [ 42 ] and a recent observational study [ 66 ] have shown that CMBTRFN also relieves psychological symptoms associated with chronic pain.

The risks and complications of CMBTRFN listed in Table 3 vary in frequency and most are minor, transient side-effects, like intraprocedural vasovagal syncope (which occurs rarely), postprocedural dysesthesia (which is relatively common), and postprocedural dizziness and ataxia (which are also relatively common). Given the prevalence of these complaints, it is prudent for patients to take precautions against falling and not drive motor vehicles until the effects have worn off. Such temporary effects are easily managed and are not likely to deter informed patients from undergoing the procedure.

Postprocedural pain occurs in virtually all cases. It varies from mild discomfort to more intense ache and lasts for varying periods of a day or so up to 2 weeks or even longer. Such pain should be relieved by analgesic medication and reassurance that it will settle.

So-called “neuritic” pain that occurs sometimes after CMBTRFN is generally more troublesome, but it varies in intensity. Two papers record neuritis, pruritis, and third occipital neuralgia as side-effects of CMBTRFN [ 3 , 13 ]. These symptoms are more commonly reported following CMBTRFN of the TON, probably due to its cutaneous innervation. The effects described are probably all on a spectrum ranging from minor irritation of a treated nerve to major inflammation causing neuralgia. They are usually of short duration, between 2 and 6 weeks, and as with other postprocedural pain, should be relieved by analgesic medication and reassurance that it will settle.

In addition to the published risks of CMBTRFN described in the Results section, there are reports in the literature of side-effects and complications of cervical RFN procedures done in ways that do not conform to the Society's guidelines. The authors felt some of those side-effects and complications should be mentioned in this review because, although never reported in association with CMBTRFN done in accordance with the Society's guidelines, they are things of which practitioners and patients considering CMBTRFN should be aware. They may be thought of as potential dangers of cervical RFN if it is done in ways that differ from that which the Society recommends.

A case of Horner's syndrome was reported as occurring after a C5-6-7 RFN procedure in 2014. The condition remitted after 36 hours with no long-term sequelae [ 67 ].

Neuritis causing third occipital neuralgia was reported as a complication in a substantial minority of patients who underwent RFN of the TON in a series published first as a meeting abstract in 2011 [ 68 ] and later as a refereed paper [ 69 ]. The problem was reported in the later paper as occurring in 19% (CI 95 9–29%) of the 64 patients treated: that figure is at odds with the numbers of severe postprocedural nerve irritation pains reported in all other studies, which record such effects as uncommon or rare. There is no certain explanation for the high incidence of neuritic pain in this series of patients, but analysis of the technique reported in the article shows that it was not the technique described in the Society's guidelines. Contrary to the guidelines, multiple electrodes were inserted at the same time. Significantly, the radiograph reproduced as Figure 3 in the article to show the first electrode placement for TON lesioning has a substantial long-axis fault, so if it had been corrected for the parallax error, the electrode would have been seen to lie much higher or lower than the operator expected. The lesion generated by that electrode may have caused inflammation of either the C2 dorsal ramus or the C3 medial branch, neither of which is meant to be a target of the procedure. Also, the article describes an additional lesion being placed deliberately “at the waist of the articular pillar of C2,” another position not described as a target site in the Society's protocol, and where a lesion is likely to irritate the C2 dorsal ramus causing lesser occipital neuralgia. All that can be said for certain is that the 19% incidence of “third occipital neuralgia” does not apply to CMBTRFN as described in the Society's guidelines.

An observational study of possible complications of cervical RFN focused on the theoretical risk of applied RF fields interfering with implanted electrical medical devices [ 70 ]. It was a retrospective review of 30 patients with implanted devices who underwent 68 RF procedures over a 5-year period. No complication had occurred and its authors concluded that RF treatment can be undertaken safely in such patients as long as precautions are observed. No other report has been published of an actual complication caused by such interference.

“Dropped head syndrome,” postoperative kyphosis due to extensor muscle denervation, has been reported twice in reports of individual patients treated by cervical RFN at multiple levels [ 71 , 72 ]. In both cases, the effects were troublesome for the patients and required treatment by surgical fusion. These reports are disconcerting. The effect seems due to denervation of cervical extensor muscles, which raises the possibility that too many levels were treated. The Society's guidelines warn excessive denervation, so it seems such outcomes can be avoided by following the guidelines.

Two cases of serious neural and vascular injuries were described in a paper in 2008 that were known to its authors by way of medicolegal proceedings [ 73 ]. The first was a patient who developed a Brown-Séquard syndrome after RF neurotomy of a C3-4 Z joint; the intraoperative films showed that electrodes were placed medial to the target joint, passing between the laminae and entering the spinal cord. The procedure was performed under general anesthesia, so the patient could not report any symptoms of impending neurological injury. The second case involved spinal cord infarction after what was supposed to be RF neurotomy of a C2-3 Z joint; films showed an electrode lying in the C3-4 intervertebral foramen (not the C2-3 foramen) and placed too far anteriorly and medially. It seems a reinforcing radicular artery was coagulated in the C3-4 intervertebral foramen, producing infarction in the territory of the anterior spinal artery.

While serious complications of neural and vascular injuries must be recognized as risks of cervical RFN, it must be stressed that they have only been described after procedures done contrary to the published guidelines. The most important safety factor stipulated in those guidelines is that the procedure should not be done under general, anesthesia; local anesthesia is all that is required. Patients undergoing CMBTRFN must be awake and fully responsive so they can report the unusual sensations that will result from inaccurate electrode placement [ 1 ]. If an electrode is placed in an incorrect position during a procedure performed under general anesthesia, as in the two cases reported above [ 73 ], the unconscious patient cannot report an unusual sensation and nontarget structures such as the spinal cord or a reinforcing radicular artery may be lesioned inadvertently, with disastrous consequences. It is too late to recognize such serious errors when the patient is woken up after a procedure but they are totally avoidable, and indeed have never been reported, when CMBTRFN has been performed on conscious patients.

The incidence of serious complications like neural and vascular injury is unknown, but they are almost certainly much more common than the two cases reported in the literature. One of the current authors knows, through medicolegal processes, of three cases of cervical lateral ramus injury with serious neurological sequelae after RF procedures performed on unconscious patients. Other authors know of similar injuries too. Clearly there is a strong bias against publication of such outcomes [ 74 ]. One reason for the under-reporting is that their publication tends to be suppressed as conditions of litigation settlement. The potential for such complications should be recognized by all who undertake CMBTRFN, but reassurance can be had from knowing that no such serious complication has ever been reported after an RF procedure performed in accordance with the Society's guidelines.

The summary of the published evidence on risks is that if CMBTRFN is performed in accordance with the Society's guidelines, only minor and temporary side-effects are likely to occur and those effects are readily managed.

The published evidence on risks of CMBTRFN and similar procedures is of low quality according to the GRADE system. Readers must be careful not to confuse “evidence of low quality” with “evidence of little significance” and perhaps go on to dismiss the risks of cervical RFN as of little concern. The evidence of risks is of low quality because it is based on observational studies and case reports. Few serious complications have been published, but the apparent publication bias, resulting in severe complications not being reported in journals, must be taken into account. Serious complications may occur if RF procedures are done in unorthodox ways, but they are extremely unlikely to occur if CMBTRFN is performed in accordance with the Society's guidelines.

Conclusions

In patients selected by comparative MBBs, fluoroscopically-guided CMBTRFN is effective for abolishing neck pain or cervicogenic headaches if done as described in the International Spine Intervention Society guidelines [ 1 ]. The procedure carries some risks, most of which are minor and temporary. Serious complications such as neural and vascular injury have been reported after cervical RFN, but they have only occurred when the Society's guidelines for the procedure have not been followed.

In accordance with the GRADE system, based on all published data on the procedure and taking into account the balance between desirable and undesirable effects, and the qualities of the evidence for each, the strength of recommendations is strong for use of CMBTRFN when the selection criteria include dual, comparative, MBBs, or placebo-controlled blocks producing 100% relief. The evidence of effectiveness of cervical RF neurotomy undertaken using any other selection criteria is unknown.

Acknowledgments

The authors wish to thank other members of the International Spine Intervention Society's Standards Division, Dr. Milton Landers, Dr. Milan Stojanovic, and Dr. Yakov Vorobeychik, who read the final draft and offered comments on it. They also wish to acknowledge Professor Nikolai Bogduk, who although not involved in the preparation of this article provided the inspiration for it. This review follows the pattern of the Society's previous systematic reviews, which he established. Also acknowledged are the invaluable contributions of Ms. Belinda Duszynski and Ms. Sandra Ray, International Spine Intervention Society staff, who assisted with managing the project, encouraged the authors’ efforts, assisted with proofreading, and helped with the reference list.

Conflicts of interest: None of the authors has any financial conflicts of interest to disclose.

References

1
International Spine Intervention Society
.
Cervical medial branch thermal radiofrequency neurotomy
. In:
Bogduk
N
, ed.
Practice Guidelines: Spinal Diagnostic and Treatment Procedures
  ,
2nd
edition.
San Francisco
:
International Spine Intervention Society
;
2013
:
165
217
.
2
Lord
SM
Barnsley
L
Wallis
BJ
McDonald
GJ
Bogduk
N
.
Percutaneous radiofrequency neurotomy for the treatment of chronic cervical zygapophysial joint pain: A randomized, double-blind controlled trial
.
N Engl J Med
 
1996
;
335
:
1721
6
.
3
Govind
J
King
W
Bailey
B
Bogduk
N
.
Radiofrequency neurotomy for the treatment of third occipital headache
.
J Neurol Neurosurg Psychiatry
 
2003
;
74
:
88
93
.
4
Acosfa
C
Grossman
RG
.
Relief of intractable pain by percutaneous anterolateral radiofrequency cordotomy
.
Tex Med
 
1960
;
65
:
36
40
.
5
Crue
BL
Todd
EM
Carregal
EJA
Kilham
O
.
Percutaneous trigeminal tractotomy. Case report: Utilizing stereotactic radiofrequency lesion
.
Bull Los Angeles Neurol Soc
 
1967
;
32
:
86
92
.
6
Shealy
CN
.
Percutaneous radiofrequency denervation of spinal facets. Treatment for chronic back pain and sciatica
.
J Neurosurg
 
1975
;
43
:
448
51
.
7
Schaerer
JP
.
Radiofrequency facet rhizotomy in the treatment of chronic neck and low back pain
.
Int Surg
 
1978
;
63
:
53
9
.
8
Schaerer
JP
.
Radiofrequency facet denervation in the treatment of persistent headache associated with chronic neck pain
.
J Neurol Orthop Surg
 
1980
;
1
:
127
30
.
9
Sluijter
ME
Koetsveld-Baart
CC
.
Interruption of pain pathways in the treatment of the cervical syndrome
.
Anaesthesia
 
1980
;
35
:
302
7
.
10
Bogduk
N
Long
DM
.
The anatomy of the so-called “articular nerves” and their relationship to facet denervation in the treatment of low-back pain
.
J Neurosurg
 
1979
;
51
:
172
7
.
11
Bogduk
N
Long
DM
.
Percutaneous lumbar medial branch neurotomy: A modification of facet denervation
.
Spine
 
1980
;
5
:
193
200
.
12
Lord
SM
Barnsley
L
Bogduk
N
.
Percutaneous radiofrequency neurotomy in the treatment of cervical zygapophysial joint pain: A caution
.
Neurosurgery
 
1995
;
36
:
732
9
.
13
Lord
SM
McDonald
GJ
Bogduk
N
.
Percutaneous radiofrequency neurotomy of the cervical medial branches: A validated treatment for cervical zygapophysial joint pain
.
Neurosurg Q
 
1988
;
8
:
288
308
.
14
McDonald
GJ
Lord
SM
Bogduk
N
.
Long-term follow-up of patients treated with radiofrequency neurotomy for chronic neck pain
.
Neurosurgery
 
1999
;
45
:
61
8
.
15
Barnsley
L
Bogduk
N
.
Medial branch blocks are specific for the diagnosis of cervical zygapophyseal joint pain
.
Reg Anesth
 
1993
;
18
:
343
50
.
16
Barnsley
L
Lord
S
Wallis
B
Bogduk
N
.
False-positive rates of cervical zygapophysial joint blocks
.
Clin J Pain
 
1993
;
9
:
124
30
.
17
Barnsley
L
Lord
S
Bogduk
N
.
Comparative local anaesthetic blocks in the diagnosis of cervical zygapophysial joint pain
.
Pain
 
1993
;
55
:
99
106
.
18
Lord
SM
Barnsley
L
Bogduk
N
.
The utility of comparative local anesthetic blocks versus placebo-controlled blocks for the diagnosis of cervical zygapophysial joint pain
.
Clin J Pain
 
1995
;
11
:
208
13
.
19
Jull
G
Bogduk
N
Marsland
A
.
The accuracy of manual diagnosis for cervical zygapophysial joint pain syndromes
.
Med J Aust
 
1988
;
148
:
233
6
.
20
Vervest
ACM
Stolker
RJ
.
The treatment of cervical pain syndromes with radiofrequency procedures
.
Pain Clin
 
1991
;
4
:
103
12
.
21
Hildebrandt
J
Argyrakis
A
.
Percutaneous nerve block of the cervical facets: A relatively new method in the treatment of chronic headache and neck pain
.
Man Med
 
1986
;
2
:
48
52
.
22
King
W
Lau
P
Lees
R
Bogduk
N
.
The validity of manual examination in assessing patients with neck pain
.
Spine J
 
2007
;
7
:
22
6
.
23
Clark
CR
Ingram
CM
El-Khoury
GY
Ehara
S
.
Radiographic evaluation of cervical spine injuries
.
Spine
 
1988
;
13
:
742
7
.
24
Jónsson
H
Jr
Bring
G
Rauschning
W
Sahlstedt
B
.
Hidden cervical spine injuries in traffic accident victims with skull fractures
.
J Spinal Disord
 
1991
;
4
:
251
63
.
25
Taylor
JR
Twomey
LT
.
Acute injuries to cervical joints. An autopsy study of neck sprain
.
Spine
 
1993
;
18
:
1115
22
.
26
El-Khoury
GY
Kathol
MH
Daniel
WW
.
Imaging of acute injuries of the cervical spine: Value of plain radiography, CT and MR imaging
.
AJR Am J Roentgenol
 
1995
;
164
:
43
50
.
27
Borchgrevink
GE
Smevik
O
Nordby
A
et al
.
MR imaging and radiography of patients with cervical hyperextension-flexion injuries after car accidents
.
Acta Radiol
 
1995
;
36
:
425
8
.
28
Barnsley
L
Lord
S
Thomas
P
et al
.
SPECT bone scans for the diagnosis of symptomatic cervical zygapophysial joints
.
Br J Rheumatol
 
1993
;
32
(Suppl 52)
.
29
Teresi
LM
Lufkin
RB
Reicher
MA
et al
.
Asymptomatic degenerative disk disease and spondylosis of the cervical spine: MR imaging
.
Radiology
 
1987
;
164
:
83
8
.
30
Boden
SD
McCowin
PR
Davis
DG
et al
.
Abnormal magnetic-resonance scans of the cervical spine in asymptomatic subjects: A prospective investigation
.
J Bone Joint Surg
 
1990
;
72A
:
1178
1184
.
31
Pettersson
K
Hildingsson
C
Toolanen
G
Fagerlund
M
Bjornebrink
J
.
MRI and neurology in acute whiplash trauma. No correlation in prospective examination of 39 cases
.
Acta Orthop Scand
 
1994
;
65
:
525
8
.
32
Voyvodic
F
Dolinis
J
Moore
VM
et al
.
MRI of car occupants with whiplash injury
.
Neuroradiology
 
1997
;
39
:
25
40
.
33
Kongsted
A
Sorensen
JS
Andersen
H
et al
.
Are early MRI findings correlated with long-lasting symptoms following whiplash injury? A prospective trial with 1-year follow-up
.
Eur Spine J
 
2008
;
17
:
996
1005
.
34
Bogduk
N
.
International Spinal Injection Society guidelines for the performance of spinal injection procedures. Part 1: Zygapophysial joint blocks
.
Clin J Pain
 
1997
;
13
:
285
302
.
35
Lord
SM
Barnsley
L
Wallis
BJ
Bogduk
N
.
Third occipital nerve headache: A prevalence study
.
J Neurol Neurosurg Psychiatry
 
1994
;
57
:
1187
90
.
36
Barnsley
L
Lord
SM
Wallis
BJ
Bogduk
N
.
The prevalence of chronic cervical zygapophysial joint pain after whiplash
.
Spine
 
1995
;
20
:
20
25
.
37
Lord
SM
Barnsley
L
Wallis
BJ
Bogduk
N
.
Chronic cervical zygapophysial joint pain after whiplash. A placebo-controlled prevalence study
.
Spine
 
1996
;
21
:
1737
44
.
38
Manchikanti
L
Singh
V
Rivera
J
Pampati
V
.
Prevalence of cervical facet joint pain in chronic neck pain
.
Pain Physician
 
2002
;
5
:
243
49
.
39
Bogduk
N
,
Editor's response, group vs categorical data in epidural studies
.
Pain Med
 
2014
;
15
:
1812
1813
.
40
Barnsley
L
.
Percutaneous radiofrequency neurotomy for chronic neck pain: Outcomes in a series of consecutive patients
.
Pain Med
 
2005
;
6
:
282
5
.
41
MacVicar
J
Borowczyk
JM
MacVicar
AM
Loughnan
BM
Bogduk
N
.
Cervical medial branch radiofrequency neurotomy in New Zealand
.
Pain Med
 
2012
;
13
:
647
54
.
42
Wallis
BJ
Lord
SM
Bogduk
N
.
Resolution of psychological distress of whiplash patients following treatment by radiofrequency neurotomy: A randomised, double-blind, placebo-controlled trial
.
Pain
 
1997
;
73
:
15
22
.
43
Balshem
H
Helfand
M
Holger
J
et al
.
GRADE guidelines: 3. Rating the quality of evidence
.
J Clin Epidemiol
 
2011
;
64
:
401
6
.
44
MacVicar
J
King
W
Landers
MH
Bogduk
N
.
The effectiveness of lumbar transforaminal injections of steroids: A comprehensive review with systematic analysis of the published data
.
Pain Med
 
2013
;
14
:
14
28
.
45
Kreiner
DS
MacVicar
J
Duszynski
B
Nampiaparampil
DE
.
The MILD procedure. A systematic review of the current literature
.
Pain Med
 
2014
;
15
:
196
205
.
46
Engel
A
King
W
MacVicar
J
.
The effectiveness and risks of fluoroscopically-guided cervical transforaminal injections of steroids: A systematic review with comprehensive analysis of the published data
.
Pain Med
 
2014
;
15
:
386
402
.
47
King
W
Ahmed
SU
Baisden
J
et al
.
Diagnosis and treatment of posterior sacroiliac complex pain: A systematic review with comprehensive analysis of the published data
.
Pain Med
 
2015
;
16
:
257
65
.
48
Bassett
K
Sibley
LM
Anton
H
Harrison
P
Kazanjian
A
.
Percutaneous Radio-Frequency Neurotomy Treatment of Chronic Cervical Pain Following Whiplash Injury: Reviewing Evidence and Needs
.
British Columbia Office of Health Technology Assessment
  ;
2001
:
40
47
.
49
Geurts
JW
van Wijk
RM
Stolker
RJ
Groen
GJ
.
Efficacy of radiofrequency procedures for the treatment of spinal pain: A systematic review of randomized clinical trials
.
Reg Anesth Pain Med
 
2001
;
26
:
394
400
.
50
Niemistö
L
Kalso
E
Malmivaara
A
Seitsalo
S
Hurri
H
.
Radiofrequency denervation for neck and back pain: A systematic review within the framework of the Cochrane Collaboration Back Review Group
.
Spine
 
2003
;
16
:
1877
88
.
51
Falco
FJE
Manchikanti
L
Datta
S
et al
.
Systematic review of the therapeutic effectiveness of cervical facet joint interventions: An update
.
Pain Physician
 
2012
;
15
:
E839
68
.
52
Uhrenholdt
L
Grunnet-Nilsson
N
Hartvigsen
K
.
Cervical spine lesions after road traffic accidents: A systematic review
.
Spine
 
2002
;
27
:
1934
41
.
53
Husted
DS
Orton
D
Schofferman
J
Kine
G
.
Effectiveness of repeated radiofrequency neurotomy for cervical facet joint pain
.
J Spinal Disord Tech
 
2008
;
21
:
406
408
.
54
Rambaransingh
B
Stanford
G
Burnham
R
.
The effect of repeated zygapophysial joint radiofrequency neurotomy on pain, disability and improvement duration
.
Pain Med
 
2010
;
11
:
1343
7
.
55
Lord SM. Cervical zygapophysial joint pain after whiplash injury: Precision diagnosis, prevalence, and evaluation of treatment by percutaneous radiofrequency neurotomy. PhD Thesis, The University of Newcastle, New South Wales, Australia, 1996.
56
Bogduk
N
Macintosh
J
Marsland
A
.
Technical limitations to the efficacy of radiofrequency neurotomy for spinal pain
.
Neurosurgery
 
1987
;
20
:
529
35
.
57
Cosman
ER
Jr
Dolensky
JR
Hoffman
RA
.
Factors that affect radiofrequency heat lesion size
.
Pain Med
 
2014
;
15
:
2020
36
.
58
Siddall
PJ
Cousins
MJ
.
Persistent pain as a disease entity: Implications for clinical management
.
Anesth Analg
 
2004
;
99
:
510
20
.
59
Cohen
M
Quintner
J
Buchanan
D
.
Is chronic pain a disease?
Pain Med
 
2013
;
14
:
1284
8
.
60
Nicholas
M
Sharp
TJ
.
Cognitive-behavioural programs: Theory and application
.
Int Anesthesiol Clin
 
1997
;
35
:
155
69
.
61
Cohen
M
Nicholas
M
Blanch
A
.
Medical assessment and management of work-related low back or neck/arm pain
.
J Occup Health Saf ANZ
 
2000
;
16
:
307
17
.
62
Vangronsveld
K
Peters
M
Goossens
M
Linton
S
Vlaeyen
J
.
Applying the fear-avoidance model to the chronic whiplash syndrome
.
Pain
 
2007
;
130
:
258
61
.
63
Engel
G
.
The need for a new medical model: A challenge for biomedicine
.
Science
 
1977
;
196
:
129
36
.
64
Smith
AD
Jull
G
Schneider
G
Frizzell
B
Hooper
RA
.
Cervical radiofrequency neurotomy reduces central hyperexcitability and improves neck movements in individuals with chronic whiplash
.
Pain Med
 
2014
;
15
:
128
41
.
65
Marshman
LAG
Kasis
A
Krishna
M
Bhatia
CK
.
Does symptom duration correlate negatively with outcome after posterior lumbar interbody fusion for chronic low back pain?
Spine
 
2010
;
35
:
657
65
.
66
Smith
AD
Jull
G
Schneider
G
et al
.
Cervical radiofrequency neurotomy reduces psychological features in individuals with chronic whiplash symptoms
.
Pain Physician
 
2014
;
17
:
265
74
.
67
Kafafy
A
Nazal
A
.
Horner syndrome after cervical median branch radiofrequency neurotomy
.
Reg Anesth Pain Med
 
2014
;
39
(
5 Suppl. 1
):
e187
.
68
Knieval
SL
Lamer
T
Pingree
M
Mauck
WD
Rho
R
.
Incidence of third occipital neuralgia following radiofrequency denervation of the C2-3 facet joint
.
Pain Med
 
2011
;
12
:
510.
69
Gazelka
HM
Knievel
S
Mauck
WD
et al
.
Incidence of neuropathic pain after radiofrequency denervation of the third occipital nerve
.
J Pain Res
 
2014
;
7
:
195
198
.
70
Barbieri
M
Bellini
M
.
Radiofrequency neurotomy for the treatment of chronic pain: Interference with implantable medical devices
.
Anaesthesiol Intensive Ther
 
2014
;
46
:
162
5
.
71
Ahmed
M
Lake
W
Resnick
D
.
Progressive severe kyphosis as a complication of multilevel cervical percutaneous facet neurotomy: A case report
.
Spine J.
 
2012
;
12
:
e5
8
.
72
Stoker
GE
Buchowski
JM
Kelly
MP
.
Dropped head syndrome after multilevel cervical radiofrequency ablation. A case report
.
J Spinal Disord Tech
 
2013
;
26
:
444
8
.
73
Bogduk
N
Dreyfuss
P
Baker
R
Yin
W
Landers
M
Hammer
M
Aprill
C
.
Complications of spinal diagnostic and treatment procedures
.
Pain Med
 
2008
;
9
(
Suppl 1
):
S11
S34
.
74
Deer
TR
.
Challenges in establishing the epidemiology of adverse events associated with interventional therapies for chronic pain
.
Pain Med
 
2008
;
9
(
Suppl 1
):
S2
S4
.