FormalPara Key Summary points

This systematic review evaluated safety and complication rates of RFA and PRF lesions of DREZC components for various pain indications.

A total of 62 manuscripts were included in this review.

Among a total of 3157 cases, there were zero serious adverse events resulting in permanent injury. A total of 36 (1.14%) transient neurological deficits, cases of transient neuritis, or non-minor adverse events like uncomplicated pneumothorax were reported. A total of 113 (3.58%) minor adverse events were reported (bruising, transient site soreness, headache).

This systematic review indicates that the use of RFA lesion of the DREZC for interventional pain management is very safe.

Introduction

Chronic pain is defined as pain that persists or recurs for longer than 3 months and is associated with significant emotional distress and/or significant functional disability [1]. The prevalence of chronic pain is approximately 25–30% of the worldwide population. It arises from numerous etiologies including tissue damage, inflammation, nerve injury, or dysfunction of the nervous system [2]. Chronic pain is a debilitating condition associated with symptoms such as unprovoked pain sensation, paresthesia, dysthesia, allodynia, or hyperalgesia [1]. A growing body of literature supports a relationship between peripheral nerve damage as well as the primary role of the plasticity and modality of dorsal root ganglia (DRG) neurons in chronic pain states [3, 4].

DRG are large collections of primary afferent sensory neurons located on the distal end of dorsal spinal roots [3]. DRG neurons are pseudo-bipolar neurons; while a peripheral neural branch innervates the target organ, a central branch carries the somatosensory information to the spinal cord where it synapses with secondary sensory neurons to transmit information to the central nervous system (CNS) [3]. The primary afferent sensory neurons are transducing information by C-fibers, A-delta, and A-beta nerve fibers from nociceptors, thermoreceptors, chemoreceptors, and proprioceptors [5]. Following peripheral nerve injury or sensitization, primary sensory neurons demonstrate maladaptive molecular changes in DRG cell bodies and in their axons, thereby resulting in development or propagation of neuropathic pain states [3,4,5]. Ion channel modifications in DRG that are in part contributing to these effects include proliferation of voltage-dependent sodium channels, downregulation of voltage-gated potassium channels, and increased expression of the calcium channels [5]. Changes occurring at the DRG neuron levels can result in peripheral sensitization, ectopic neuronal activity, presynaptic modulation as well as increased neuronal excitability in the spinal cord [5]. Further, hyperexcitability and ectopic firing of cell bodies at the DRG level can increase central sensitization and reduce central inhibition that are crucial for the onset and maintenance of chronic pain [5].

The critical role of the DRG neurons in pain transduction and preservation of persistent pain states has long been recognized, focusing attention on these tissues as targets for therapeutic interventions [4]. The dorsal sensory pathway’s linear anatomical organization and accessible location have made the DRG and the neighboring dorsal rootlets, dorsal root entry zone, as well as the dorsal rami and their branches, which are collectively referred to as the dorsal root entry zone complex (DREZC), an attractive target for neuromodulation [4, 6]. DREZC components, which are part of the dorsal sensory pathway, have been the target of pain relief interventions including radiofrequency ablation (RFA) and steroid injections [4,5,6,7,8]. Biological changes in neurological tissues exposed to RFA can occur as a result of the thermal effects and/or the high intensity electric field, and can be either ablative or neuromodulatory lesions [4,5,6,7,8].

While the DRG, being central in pain signal processing and propagation of chronic pain states, is a logical target component of the DREZC, it is technically difficult to accurately discern whether the tips of each RFA cannula are adjacent to the DRG or adjacent its neighboring DREZC components, since the DRG cannot be visualized using fluoroscopy, the most commonly used modality for RFA procedures [6]. DRGs also vary in number, from one to three per spinal level, and in location (intraspinal, intraforaminal, or extraforaminal), none of which can be visualized or reliably identified using fluoroscopy [6].

As such, there is an element of potential error in all past publications reporting interventions of the DRG, specifically. As a result of the anatomical differences of individualized components of this linear dorsal afferent sensory pathway (the dorsal rami, DRG, the dorsal rootlets, and the dorsal root entry zone) and limitations to accurately localize each of these components with common clinically available tools (such as fluoroscopy), the more sensitive and inclusive term DREZC has been utilized herein to refer to RFA lesions targeting the DRG or dorsal rami [6]. The goal of this systematic review is, thus, to assess the clinical safety and potential complications of conventional and pulsed RFA targeting DREZC components in the treatment of chronic pain.

Methods

Study Design

Systematic literature review, PRISMA method.

Search Strategy

This systematic search was limited to peer-reviewed literature using “radiofrequency ablation” as a search keyword using PubMed’s database for manuscripts published between inception and December 2020.

Study Selection

Abstracts that involved the application of radiofrequency currents, of any modality, to DREZC components for the treatment of pain were included for full-text review.

Inclusion Criteria

Search was limited to original data describing clinical outcomes following RFA performed for pain indications only, involving the DREZC components outlined above, in human subjects, and written in English.

Exclusion Criteria

Exclusion criteria included the following:

  • Technical reports or basic science investigations not describing clinical outcomes following RFA procedures

  • Manuscripts describing RFA not targeting DREZC components

  • Manuscripts describing the use of RFA for non-pain indications (i.e., cardiac, dermatology, cancer)

  • Review articles

  • Case reports

Data Collection

Two trained research assistants employed by author OV performed the initial identification and screening, confirming each other’s findings. All authors subsequently assessed these reports for eligibility and inclusion, and obtained confirmation from at least one colleague as to the eligibility of each manuscript.

Data Extraction

Data was extracted from these reports into a spreadsheet by the two trained research assistants employed by author OV, then subsequently verified by authors OV and MP. The primary outcomes were complications, which were subcategorized into the following:

  • Type 1 = Persistent neurological deficits or other serious adverse events, defined as any event that resulted in permanent of prolonged injury.

  • Type 2 = Transient neuritis or neurological deficits, or other non-neurological non-minor adverse event.

  • Type 3 = Minor adverse events (e.g., headache, soreness, bruising, etc.).

Variables Measured

The following parameters were measured:

  • Indication for DREZC ablation

  • Type of radiofrequency ablation (conventional or pulsed)

  • Location of the treatment (anatomic targets)

  • Complications associated with the treatment

  • Duration of the complications associated with the procedure

Effect Measures and Synthesis of Data

Data for the above complications were tabulated to note prevalence rates of each type of complication (Table 1), but no quantification of results was planned as it was expected there would be significant heterogeneity in the data. Missing data were noted as such.

Table 1 Summary of complications following conventional or pulsed RFA of the dorsal root entry zone complex (DREZC)
Full size table

Reporting Bias

No attempts were made to assess risk of bias due to missing results arising from reporting biases as the authors felt such efforts would be largely subjective.

Ethics Compliance

This article is based on previously conducted studies and does not contain any new studies with human participants or animals performed by any of the authors.

Results

A total of 39,207 manuscripts from PubMed published between inception and December 2020 were identified in the initial literature search and the majority were excluded (Fig. 1). A total of 62 manuscripts were included for qualitative analysis: 25 were randomized controlled trial (RCT) studies, 16 were retrospective studies, 19 were prospective studies, and two were quasi experimental studies, collectively reporting outcomes for a total of 3157 patients. The characteristics of the included studies are detailed in Tables 2, 3, 4, and 5.

Fig. 1
figure 1

Adapted from Preferred Reporting Items for Systematic Reviews and Meta-Analyses [PRISMA]; Moher et al., 2009. RFA = radiofrequency ablation; Pain other = manuscripts not describing clinical use of RFA, but describing some other aspect of pain pathophysiology or clinical pain management

Flow diagram of study selection.

Full size image

Randomized Controlled Clinical Trials

Twenty-four RCT studies were analyzed, including nine RCTs using conventional RFA and 15 studies with pulsed mode RFA (Table 2) [9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32]. Of the nine conventional RFA RCTs, two pertained to cervicobrachial pain, one pertained to cervicogenic headache, one pertained to lumbosacral radicular pain, one pertained to thoracic radicular pain, and four pertained to chronic low back pain or lumbar facet joint pain. A total of 402 patients were exposed to conventional RFA between 40 and 80 °C. Two studies for cervicobrachial pain showed that RFA lesions adjacent to the DREZC were associated with mild complications post treatment [9, 10]. In a randomized trial that exposed nine patients to conventional RFA at 67 °C, seven patients complained of a burning sensation in the treated dermatomes that receded spontaneously after 3 weeks [10]. In the same study, one patient also had mild hypoesthesia of the upper arm which decreased after 3 months [10]. Another randomized study for cervicobrachial pain reported complaints of neuritis and a slight loss of muscle strength in a small number of patients at 6 weeks post treatment that resolved spontaneously during a 3-month follow-up period [9]. Interestingly, an RCT trial that implemented conventional RFA for cervicogenic headache in 15 patients was not associated with any serious complications [11].

Table 2 Randomized controlled clinical trial data: complications following conventional or pulsed RFA of the dorsal root entry zone complex (DREZC)
Full size table

A randomized, double blind, sham-controlled study in 45 patients using conventional RFA for chronic lumbosacral radical pain reported discrete loss of motor function and change in sensation in 29 patients [12]. Another study that described the effect of conventional RFA of the DREZC in 78 patients suffering from thoracic refractory pain due to chest malignancies also reported adverse events post therapy, including back pain (n = 9), soreness (n = 15), and hematoma (n = 3) whereas major complications were neuritis (n = 11), sensory deficit (n = 8), and anesthesia dolorosa (n = 2) [13]. Notably, however, in this study, RFA of thoracic T2–T8 DREZC lesions with combined fluoroscopy and CT-guidance that was applied to 40 out of 78 patients showed significantly lower incidence of adverse events in comparison to patients that were treated with standard fluoroscopy (15% vs. 37%, respectively) [13]. No infection, motor deficits, or pneumothorax was recorded in this clinical trial [13]. In an RCT that treated 15 patients with conventional RFA at the dorsal rami (components of the DREZC) of L3, L4, and L5 for chronic low back pain, no complications associated with the procedure were reported [14]. In another trial with conventional RFA on dorsal ramus for low back pain, two of 151 treated patients experienced significant worsening of back pain while one patient had a new radiating pain in their leg after the procedure [15]. All symptoms in affected patients spontaneously resolved after 3 months [15]. In a third RCT using conventional RFA on the dorsal ramus to treat chronic low back pain in 56 patients, no adverse events were reported during the 6-month observation period [16]. Similarly, the application of conventional RFA to the medial branch of the dorsal ramus to 30 patients with lumbar facet joint pain was not encountered with serious adverse events during the trial [17].

Of the 15 published RCTs with pulsed RFA of DREZC components (Table 2), three studies were for patients with cervical pain syndromes [18,19,20]. In one RCT that implemented pulsed RFA in 11 patients with chronic cervicobrachial pain, no treatment-associated complications were observed [18]. In a smaller RCT study, five patients treated with pulsed RFA for the reduction of radicular pain in cervical and lumbar dermatomes also did not report complications or adverse events during the extended follow-up period [19]. Another RCT study in 34 patients with cervical disk herniation was associated with mild and transient headache as well as muscle stiffness in some patients post procedure [20].

There were also nine RCTs of pulsed RFA lesions to DREZC components in patients with chronic lumbar or lumbosacral pain [21,22,23,24,25,26,27,28,29]. One RCT that evaluated pulsed RFA and electroacupuncture on low back pain did not report any associated complications over 1 month after therapy in 100 patients [21]. Another RCT including 30 patients with chronic lumbar radicular pain treated with pulsed RFA found no serious short-term and long-term side effects, but did note two of 16 patients treated with RFA complained of minor headache and a transient increase in back pain which did not last beyond 1 day [22]. In an RCT study that used three cycles of pulsed RFA of the lumbar DREZC in 31 patients with chronic lumbosacral pain, several patients reported temporary pain during needle insertion and paresthesia during sensory stimulation as well as pain aggravation at 2–3 days post procedure that resolved spontaneously without any sequelae; however, no serious adverse events were noted [23]. Similarly, pulsed RFA on the dorsal ramus, a component of the DREZC, on 55 patients with lumbar facet joint pain reported transient pain (up to 2 days) post procedure [24]. An RCT with a single treatment of RFA on lumbar DREZC in 11 patients reported no treatment-related side effects [25]. Another RCT study that also used pulsed RFA lesions of the DREZC in the treatment of low back pain in 60 patients had no associated complications [26]. A study in 25 patients with lumbar radicular pain that combined transforaminal epidural bupivacaine injection with three cycles of pulsed RFA did not report complications with either treatment modality [27]. An RCT that tested effectiveness of combining pulsed RFA on lumbosacral DREZCs with epidural adhesiolysis showed that pulsed RFA application for the neuropathic pain due to lumbosacral radiopathy was not associated with any complications [28]. An RCT study of 150 patients with chronic lumbar facet syndrome treated with pulsed RFA of the DREZC or radiofrequency denervation of the medial dorsal branch also noted no complications arising as a result of either treatment [29].

There was one study that examined RCT of pulsed RFA lesions of thoracic DREZCs for postmastectomy pain syndrome [30]. Analgesic efficacy of pulsed RFA lesions was evaluated on the DREZC and compared to RFA of their corresponding paravertebral somatic nerves. Both RFA treatments had no side effects; nevertheless the authors acknowledged the inherent risk of performing thoracic foraminal interventions and the technical difficulty of targeting thoracic DREZC components [30].

The efficacy of pulsed RFA was also tested on herpes zoster neuralgia in two RCT studies [31, 32]. A study in 150 patients with acute, subacute, and chronic herpes zoster neuralgia targeted the DREZC with CT-guided pulsed RFA [31]. Two cases of uncomplicated pneumothorax were observed among 150 patients [31]. The second RCT for the post-herpetic neuralgia in 168 patients of which 58 were exposed to pulsed RFA combined with gabapentin had no associated complications [32].

Retrospective Studies

There were 15 studies with a retrospective design that used pulsed RFA and one study that evaluated the combination of pulsed and conventional RFA (Table 3) [33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48]. Of the five retrospective studies of pulsed RFA on cervical DREZCs, two studies evaluated the treatment on cervicogenic headache and three studies evaluated the treatment of chronic cervical pain syndromes [33,34,35,36,37]. A case–control study that evaluated 139 patients of which 87 patients had RFA in combination with epidural steroid injection at the C2 level for cervicogenic headache had no complications associated with the RFA intervention [33]. In another retrospective cervicogenic headache analysis of 45 patients that received RFA lesions to DREZC components of C2, no post-procedure complications were recorded throughout the study [34]. In a retrospective study of 59 patients who underwent RFA therapy for chronic cervical radicular pain, a single patient experienced a pain flare-up post procedure that spontaneously resolved after 2 weeks [35] while a retrospective review of 18 patients who underwent RFA for chronic cervicobrachial pain did not show any adverse events [36]. Likewise, a study of 154 patients treated with pulsed RFA as a result of cervical and lumbar radicular pain did not report any complications [37].

Table 3 Retrospective studies: complications following conventional or pulsed RFA of the dorsal root entry zone complex (DREZC)
Full size table

In a retrospective data analysis of 279 patients who received RFA for chronic lumbar radicular pain, the authors did not report complications associated with RFA for lumbar radicular pain in any of the treated patients [38]. Another study in 13 patients with acute lumbar radicular pain who were possible candidates for disk surgery, RFA of the DREZC was not associated with any side effects up to 12 months post treatment [39]. In another study that examined RFA in 60 patients with lumbosacral radicular pain syndrome, no complications were reported [40]. A retrospective analysis of 50 patients that were exposed to multiple conventional and pulsed RFA treatments showed a single adverse event of transient thigh numbness which resolved after 1 week [41], whereas a study that retrospectively analyzed 60 patients with chronic lumbosacral radicular pain had no complications related to pulsed RFA [42]. In a retrospective study that included 82 patients with intractable lumbosacral radicular pain who had poor clinical outcomes after lumbar spinal surgery and subsequently underwent pulsed RFA of the DREZC, the authors did not report any complications associated with the RFA procedure [43]. In another retrospective study that analyzed 28 patients with neuropathic spinal pain who had pulsed RFA rhizotomy of the DREZC no major complications were noted. However, in this study a small number of patients reported mild discomfort in the treated area that resolved spontaneously within 3 weeks [44]. Twenty-two patients with persistent postsurgical pain after knee arthroplasty who received pulsed RFA of the L4 DREZC in combination with transcutaneous electrical nerve stimulation reported no associated complications [45].

In a retrospective study that evaluated 49 patients with chronic postsurgical thoracic pain, 15 patients who received pulsed RFA on the intercostal nerves and 13 patients who had of DREZC RFA were compared with 21 patients who were treated pharmacologically [46]. One patient had an adverse event of a small pneumothorax that was detected during a routine scan after pulsed RFA of the DREZC [46]. The pneumothorax was found to not be related to the procedure and was treated conventionally. The second case of pneumothorax required hospitalization and was reported in a patient that was treated with RFA on the intercostal nerves [46].

There were two retrospective studies of pulsed RFA of the DREZC for the management of acute and post-herpetic herpes zoster neuralgia [47, 48]. The first study examined 42 patients, 22 that received continuous epidural block and 20 who were treated with pulsed RFA of the DREZC [47]. Only one patient in the pulsed RFA treatment group had pain at the site of the procedure [47]. The second study evaluated 58 patients who were treated with pulsed RFA of the DREZC either at the acute herpes zoster or post-herpetic neuralgia: no complications were reported [48].

Prospective Studies

There were 18 prospective design studies of which eight studies were conventional RFA, eight were pulsed RFA, and two studies used a combination of pulsed RFA with conventional RFA modalities (Table 4) [49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66].

Table 4 Prospective studies: complications following conventional or pulsed RFA of the dorsal root entry zone complex (DREZC)
Full size table

Four prospective studies applied conventional RFA lesions to cervical DREZCs [49,50,51,52]. The first study followed 20 patients for 6 months and 17 patients for 9 months post DREZC lesioning [49]. A total of 12 patients experienced burning pain whereas seven patients had hyposensibility in the dermatome, both of which resolved within 3 weeks post treatment [49]. Additionally, one patient experienced prolonged hyposensibility up to 6 months post treatment [49]. The second prospective study that treated cervicogenic headache with conventional RFA on the dorsal ramus reported that one of 15 patients had burning pain post procedure which resolved after 2 weeks [50]. The third study in 54 patients with cervicobrachialgia did not have complications post procedure [51]. The fourth study treated 122 patients with conventional RFA on the dorsal ramus of either in the thoracolumbar region or cervical spine [52]. There were no major complications associated with the procedure; however, 27 patients reported discomfort at the site of operation which resolved spontaneously after 1 month [52].

A conventional RFA study in 17 patients with hip flexor/adductor spasms and pain from cerebral palsy found that two patients had an increase in pain post procedure, which was temporary and treated with gabapentin [53].

Another prospective study that used conventional CT-guided RFA in patients with refractory pain induced by rib metastasis of lung cancer reported that 15 out of 27 patients had chest wall numbness at 3 days post procedure and 12 patients had chest wall numbness both at 1 and 3 months post procedure [54].

Interestingly, a study that reported on 58 elderly patients, aged 80 and older, treated with conventional RFA on the dorsal ramus, a DREZC component, for low back pain did not have any major complications post procedure, but the authors noted that 11 patients developed discomfort at the site of intervention that resolved spontaneously by 1-month follow-up [55]. Another study that also used conventional RFA on the dorsal ramus to treat chronic thoracic spinal pain reported transient postoperative pain in five of the 40 patients [56].

In a study that treated 76 patients for chronic lumbar radicular pain, 37 of which had pulsed RFA and 39 who had pulsed RFA and conventional RFA, no neurological deficits were observed in either treatment group [57]. In a study that treated 118 patients with lumbar facet joint pain, of which 75 patients had pulsed RFA and 43 patients had conventional RFA on the DREZC [58], the authors reported that two patients treated with pulsed RFA developed neuropathic pain after three repetitions while one patient treated with conventional RFA developed neuropathic pain after two repetitions [58].

Of the eight pulsed RFA manuscripts, one pertained to chronic headache, two pertained to cervical radicular pain, four pertained to lumbosacral radicular pain, and one pertained to post-herpetic neuralgia [59,60,61,62,63,64,65,66]. In 20 patients suffering from chronic headaches treatment with ultrasound-guided pulsed RFA was not associated with any major complications, but the authors reported that one patient had transient cervicalgia which resolved within 24 h and three cases of transient dizziness [59]. In another study cervical DREZC RFA, none of the 15 patients exposed to RFA experienced any side effects following treatment [60]. Similarly, treatment with bipolar pulsed RFA for cervical radicular pain was not related to any complications in 20 treated patients [61].

A total of four prospective studies evaluated pulsed RFA intervention in 232 cumulative patients for the treatment of chronic lumbar and lumbosacral radicular pain [62,63,64,65]. Pulsed RFA DREZC lesions were performed from L2 to S1 levels and there were no complications reported in any of the four manuscripts [62,63,64,65].

Similarly, in a prospective study that evaluated three cycles of pulsed RFA in 49 patients for the treatment of post-herpetic neuralgia, the authors did not report any associated complications [66]. Another prospective study involving 90 patients with post-herpetic neuralgia who were treated with high voltage, long duration, bipolar pulsed RFA, the side effects included high blood pressure, pain, and tachycardia [67]. Notable bruising was present in five patients post treatment that spontaneously recovered during the follow-up period [67].

Quasi and Controlled Clinical Studies

There were two quasi studies and one controlled clinical study (Table 5) [68, 69]. Both quasi studies focused on lumbar radicular pain, one of which combined pulsed RFA and conventional RFA, whereas the second study only used pulsed RFA as the treatment modality. The first quasi study treated 25 patients with pulsed RFA, conventional RFA, or a combination of pulsed RFA and conventional RFA [68]. No major complications were reported; however, a few patients experienced mild pain at the site of puncture post procedure, which spontaneously resolved after several days [68]. In the second study 25 patients were exposed to pulsed RFA and had no reported complications [69].

Table 5 Quasi and controlled clinical studies: complications following conventional or pulsed RFA of the dorsal root entry zone complex (DREZC)
Full size table

Discussion

Given the significance of the DREZC components in the development and propagation of chronic pain, RFA modalities have been recognized as a clinically important approach for interventional pain management. RFA lesioning of the DREZC components is a minimally invasive procedure with the potential to afford more permanent and complete denervation. Of the 62 selected manuscripts totaling 3157 patients, there were zero serious adverse events or persistent neurological deficits reported. A total of 36 (1.14%) transient neurological deficits, cases of transient neuritis, or non-minor adverse events like uncomplicated pneumothorax were reported. A total of 113 (3.58%) minor adverse events were reported (bruising, transient site soreness, headache).

Initial clinical investigations have shown that conventional RFA could be used safely with low incidence of post-procedural neuropathic pain. In recent years, pulsed RFA treatment administered to DREZC has generated compelling evidence of its efficacy in clinical practice for a variety of pain syndromes (Tables 2, 3, 4, and 5). Although conventional and pulsed RFA on DREZC are deemed distinct, both techniques were applied to treat similar pain syndromes including cervicogenic headaches, cervical radicular pain, discogenic pain, lumbar radicular pain, and pain associated with the sacroiliac joint [4, 7, 8, 11, 12, 19, 20, 22, 23, 27, 28, 33,34,35, 37, 39,40,41,42,43, 50, 57, 59,60,61,62,63,64, 67, 68]. Further, RFA treatment strategy has been shown to be beneficial in treating disease states such as acute and post-herpetic neuralgia and refractory pains post cancer surgery [13, 30,31,32, 47, 54].

RFA is a minimally invasive treatment option with good results for a variety of pain states and, herein, we have demonstrated that it has an excellent safety profile. Of the few adverse events, most reported only minor procedure site-related adverse events, like soreness or bruising. There were, however, two cases of pneumothorax reported among 150 patients treated for herpes zoster neuralgia [31], but these were considered uncomplicated and not hemodynamically compromising.

Notably, it is important to note that the studies selected for analysis were not designed with the primary outcome of safety and, thus, were not powered to detect complications of RFA lesions of the DREZC. Nonetheless, collectively, among the 3157 patients in the 62 analyzed manuscripts, no serious adverse events manifested.

Conclusion

This systematic review indicates that RFA lesions of the DREZC for interventional pain management are quite safe. There were no serious adverse effects with a sizable sample of RCT, prospective observational, and retrospective studies.