INTRODUCTION
Low back pain is a leading cause of disability and economic impact in developed countries [1,2] and management of axial low back pain from the anterior column remains a key challenge. Further, the role of procedural and surgical treatment of axial low back remains controversial [3-5]. A subset of patients with axial low back pain will have vertebrogenic pain (VBP) as evidenced by type 1 and 2 Modic changes on magnetic resonance imaging (MRI), which have a prevalence between 5.8%–13% depending on the populations studied [6,7].
While the level of evidence behind many low back pain treatments have been questioned, basivertebral nerve ablation (BVNA) targeting VBP linked to type I and II Modic changes has support as a technique from a randomized double-blind sham-controlled trial [8], evidence of procedural efficacy compared to standard care [9] and subsequent reproduction of positive efficacy in ‘real world’ prospective studies [10].
Among the key structures implicated in the pathophysiology of VBP is the basivertebral nerve (BVN) plexus, which is a nociceptive plexus that innervates the vertebral endplates (VEPs) and is thought to play a central role in the development of pain originating from the vertebral body. The BVN arises from paired branches of the sinuvertebral nerve and supplies the caudal and cranial aspects of the VEP, which itself has limited innervation from other sources, including the grey ramus communicans and additional penetrating branches from the sinuvertebral nerve as demonstrated in Figure 1 [11]. Damage or degeneration to the VEP, as demonstrated by types 1 and 2 Modic changes on MRI or increased uptake on single photon emission computed tomography/computed tomography (SPECT/CT) bone scan, in the presence of conditions like degenerative disc disease (DDD), can lead to an increase in nociceptive signalling and the emergence of VBP [12]. Studies have demonstrated that the BVN plexus harbors a higher density of nociceptors in damaged endplates compared to normal intact endplates, underscoring its potential role in pain generation [13].
Recent advances have highlighted the BVN as a potential target for radiofrequency (RF) ablation. Techniques such as BVNA using RF energy delivered via a transpedicular approach have shown promise in selectively interrupting pain pathways without causing significant damage to adjacent tissues. The original safety data of intravertebral RF ablation was presented in 2012 [14]. The aforementioned study tested a novel multitined expandable RF electrode on Elk spines to assess its safety and efficacy for treating spinal pain, particularly in metastatic disease (Figure 2A and B). The electrode produced stable, predictable lesions without causing unwanted thermal spread beyond the vertebral body, suggesting its potential for safe use in intervertebral tumor ablation. The first case presentation of this bipedicular BVNA technique for the management of lower back pain was in 2016 [15], with significant reduction in pain severity at 12-month follow-up. The Intracept device (Boston Scientific Corp., USA), a patented unipolar transpedicular system, has since emerged as the most well-studied technology for BVNA. Randomized controlled trials (RCTs) have demonstrated its excellent safety profile, as well as providing both short-term and long-term pain relief for the majority of patients with VBP [13].
Despite the promising results in treating VBP, the use of BVNA has been limited primarily to patients with Modic type 1 and 2 endplate degeneration. Emerging literature, however, has suggested that BVNA may have broader applications. Recent case series have reported successful treatment of VBP in patients with conditions such as tumor-related vertebral pain, spinal deformities prior to fusion surgery, and adult spinal deformity patients undergoing multimodal surgical interventions [12,16]. This expanding body of evidence suggests that BVNA may offer a novel therapeutic avenue for a variety of VBP syndromes, particularly in cases where traditional conservative pain management strategies have proven ineffective, and surgery is often considered the only option.
The aim of this paper is to present early experience with a technique for BVNA performed via bilateral transpedicular access using pedicle access needles, and to review the literature on the safety, efficacy, and expanding indications for this procedure. This cohort study explores the outcomes of this technique in a cohort of patients and aims to contribute to the growing body of evidence supporting the use of BVNA in expanded indications beyond the treatment of degenerative endplate pain.
MATERIALS AND METHODS
1. Design
A prospective cohort study was performed in line with the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines [17]. Ethics approval was gained for study into subjects with spinal pain who had consented to participation through Bond University Human Research Ethics Committee upon outcomes of interventions for patients with spinal pain - reference MZ00044.
2. Participants
Multicentre recruitment took place in a private spine clinic and interventional radiology practice in the Gold Coast, Queensland, Australia. Participants were included on the premise of chronic midline axial lower back pain of sharp description aggravated by activity and flexion. Pain was not aggravated by extension, nor neuropathic in nature, and evidence of type 1 or 2 Modic changes must be present or, if absent, increased endplate uptake should be seen on Technetium Bone Scan/SPECT CT. These criteria, with the exception of the nuclear medicine bone scan results, are based off prior descriptions by Fischgrund et al. and the Intracept protocol (Boston Scientific Corp.) [13,18]. Hounsfield units (HU) were calculated from region of interest cancellous vertebral bodies, where the patient had a CT within 12 months prior to the procedure. The mean average of 2 measurements for each level within the vertebral body was utilised and targeted (e.g., L4–5 level had 2 measurements for L4 and L5 respectively). Participants were excluded on the basis of significant instability specified as unstable fractures and over grade 2 spondylolisthesis. For patients with vertebral lesions, such as stable fractures or tumors, only the affected vertebra would undergo BVN ablation, while other vertebrae showing signs of DDD or axial mechanical spine disease may be treated if Modic changes or bone scan uptake are present. A spinal surgeon (MZ) assessed all patients for suitability, performing most of the interventions in theatre and referring 3 patients for radiology interventions with 2-month follow-up to assess progress.
3. Equipment and Procedure – Theatre Based Intervention
The procedure was conducted in accordance with Wright’s seminal publication of 2016 [15]. The procedures were performed by a spinal surgeon (MZ) trained in the BVNA technique with the patient prone under conscious sedation. Under fluoroscopic guidance, an 11- or 13-gauge introducer was advanced to a position near the juncture of the dorsal and medial one third of the vertebral body as viewed from lateral with fluoroscopic imaging (Figures 3 and 4). After insertion of the introducer and confirmation of depth on lateral, the fluoroscopy beam was then returned to the initial posterior oblique angle to visualise down the introducer. Once positioned, the introducer was removed and pediculotomy visualised. The Nimbus (RF needle; Stratus Medical LLC, USA) was advanced into the pediculotomy to the vertebral body. Once both cannulas were inserted, lateral imaging was used to adjust depth of both cannulas to the juncture of the dorsal and medial one third of the vertebra. Contrast should be injected to exclude medial pedicle breach during placement. The 17-gauge Nimbus RF needle (Figure 5) was positioned to bracket the described nexus of the BVN as described by Antonacci et al. [19] and Bailey et al. [20]. While optimal interelectrode distance is less than 20 mm [21], larger distances are feasible for lesioning due to the high electrical and thermal conductance of intravertebral trabecular bone marrow. Thermal concordance was readily established between the 2 cannulas with the cathode set to 80°C as observed by temperatures at the anode which will exceed 65°C. With technically correct performance, postprocedure MRI demonstrates 2 discrete foci of intense intravertebral ablation with clear changes to the trabecular bone 'stippling' (T2) between the cannulas. Notably the BVN is a continuation of the sympathetic sinuvertebral nerve and can be successfully ablated at temperatures below those required to fully coagulate peripheral somatic nerves. A unipolar approach was considered in the case of an unsuitable pedicle secondary to sclerotic entry point or prior pediculoplasty. Following fluoroscopic confirmation of positioning, bipolar thermal ablation was delivered for 5–6 minutes at 70°C to 80°C.
4. Equipment and Procedure – Radiology CT Suite
The procedures were performed by a musculoskeletal interventional radiologist (CAB) with over 15 years of image-guided RF ablation experience. Patients were placed prone on the CT gantry and limited scout scan and low dose helical scan of the treatment level performed to assess gantry tilt requirements. With the patient fasted, conscious intravenous sedation was administered. Using gantry tilt of 15°–30° depending on the level treated, transpedicular access with 11- to-13-gauge Jamshidi needles was achieved under CT fluoroscopy guidance. To optimise lesioning, a 17-gauge Nimbus 150-mm MEE (multitined expandable electrode) RF cannula was positioned aiming for the probe tips to be at the junction of the middle and posterior one third of the vertebral body in the sagittal plane. Bipolar thermal RF ablation was performed between 85°C to 90°C for 6–8 minutes. Reproduction of usual VBP during the procedure was confirmed with patient questioning. Patients were discharged the same day, with recommendation for reduced activities for 3–5 days and oral analgesia as required for 5- to 7-day postprocedure and followed up with the referring spinal surgeon.
5. Data Collection
Data was collected and stored on an encrypted database and provided in a deidentified fashion for analysis. Baseline data including patient demographics and indications were recorded. Satisfaction at a two-month review was recorded using a Likert scale and visual analogue score (VAS), along with any complications.
RESULTS
1. Participant Demographics
Forty patients, comprising of 24 males (60%) and 16 females (40%) with a mean age of 63.92 years and 65.56 years, respectively participated in the study (Table 1). Indications for the procedure divided by sex are provided in Table 1, with degenerative type 1 and 2 Modic MRI endplate signal comprising of the majority of cases across both genders (80% total). Males were more represented in degenerative type 1 and 2 Modic MRI endplate signal, while females and males were equally predominant in the other indications compared to males.
2. Treatment Results Analysis
All participants underwent the BVNA procedure with no complications. Satisfaction measured by the Likert scale at 2-month review can be seen in Table 2, with 75% of participants rating 4 or 5. There was no significant difference in outcomes between genders as per the Student t-test (p=0.593) (Table 3). The lowest satisfaction rating of 1 was by a 79-year-old female with scoliosis who was referred from a pain physician on the background of a previous failed spinal cord stimulator, high central sensitisation on the screening questionnaire and opioid use.
3. Expanded Indications Analysis
A total of 8 participants were included for nondegenerative Modic type 1 and 2 endplate change indications including; adjacent multilevel segment disease (AMSD), postfracture, inflammatory endplate osteitis, myeloma deposition/fracture and hemangioma. There was no significant difference in satisfaction between participants included for Modic type 1 or 2 endplate changes compared to the expanded indications (p=0.366) (Table 3). As summarised in Table 3, the change in VAS scores from pre- to postoperatively was 2.97±2.16 in participants with Modic type 1 or 2 endplate changes, compared to 3.63±2.07 in those treated under expanded indications, a difference that did not reach statistical significance (p=0.444). Both female participants included for AMSD improved in their symptoms, reporting 100% and 50% relief. Two male patients with postfracture pain reported 40% and 70% improvement in symptoms respectively. Participants included for inflammatory and hemangioma had a complete resolution of symptoms, while the postmyeloma participant had 30% reduction along with a 50% reduction in opioid doses and sleep improvement.
DISCUSSION
This report details the application of bilateral transpedicular BVNA for treatment of VBP of varying indications. The primary outcome measure of patient satisfaction (Likert rating) for broadened indications had no significant difference compared to the Modic type 1 or 2 group or 'Group 1'. In the expanded indications analysis, 8 patients were included in 'Group 2' which comprised of AMSD, postfracture, postmyeloma infiltration/fracture, inflammatory spondyloarthropathy and hemangioma. The secondary outcome measure of pain relief was also significant, with many patients within the ‘expanded indications’ group reporting between 40% to 100% pain relief and improvement. General patient satisfaction for all who underwent BVNA was clinically significant, with 75% of subjects reporting a rating of ‘4’ or ‘5’ at 2-month review, as seen in Table 2.
Concerning the etiopathogenesis of chronic lower back pain, numerous studies have postulated that certain patients may have a vertebrogenic component rather than exclusively a discogenic one which is commonly the assumed cause in suspected anterior column pain. This is likely due to VEPs being highly vascularised and innervated compared to the intervertebral discs and ligamentous structures and thus potentially nociceptive to local pathology. In addition to a thorough history and physical exam detailing whether mechanical back pain is worse with load and flexion, the presence of Modic changes in MRI is a successful predictor of the contribution from VBP [16]. Established through immunohistochemical staining, the BVN contains pain-associated neurotransmitters (substance P, CGRP [calcitonin gene-related peptide], and Trk-A), emphasising its role in pain transmission [22]. This evidence supports the therapeutic claim that the interruption of afferent vertebrogenic signalling could represent a new and effective alternative for treating not just chronic lower back pain, but further indications beyond traditional degenerative conditions. The seminal study by Becker et al. [23], marked a pivotal shift in the understanding of chronic low back pain by identifying the VEPs as a principal pain source in a distinct subset of patients. Their findings demonstrated that BVNA is a safe, minimally invasive, and tissue-sparing intervention that significantly reduces disability, with a 57.1% improvement in Oswestry Disability Index (ODI) scores at 1 year. Importantly, the procedure offers durable pain relief, is suitable in the early stages of treatment, and preserves future therapeutic and surgical options – establishing a new treatment paradigm for VBP.
Reported indications for BVNA from previous publications of large-scale studies include those patients with axial chronic lower back pain of vertebral aetiology of at least 6-month duration who have failed conservative treatment and exhibit radiological evidence of Modic type 1 or 2 changes at the VEPs of the levels targeted for treatment [16,24]. Conversely, multicentre studies such as Fischgrund et al. [24], omit patients with diagnosed osteoporosis, spondylolisthesis and symptomatic spinal stenosis.
There is limited evidence in literature for the use of BVNA in alternate indications such as a pathologic fracture, hemangioma or AMSD cases. In an American study by Fogel et al. [12] investigating the effectiveness in spondylolisthesis and scoliosis, BVNA proved reliable in reducing VAS and ODI scores in patients with or without significant comorbidities. This study broadens the scope of indications for BVNA beyond degenerative Modic type 1 and 2 MRI endplate changes, highlighting the potential relevance of conditions commonly encountered in clinical practice—such as spinal deformities and prior spinal surgery—which have been systematically excluded from previous Level I randomized controlled trials. Additionally, vertebral endplate irregularities caused by hemangiomas and Schmorl nodes do not appear to interfere with BVNA targeting and, as such, warrant further investigation [25].
BVNA has also been proposed for patients where surgical fusion is considered not only for pain relief but for use as a modality to predict further open surgical treatment. It may also be of importance in limiting the degree of lumbar fusion ultimately required by treating or excluding levels that would otherwise be included in the reconstruction [12]. In a double-blinded RCT evaluating opioid use in chronic lower back pain, patients who underwent BVNA with decreased opioid use had significant improvements in ODI and VAS scores compared to those reporting increased opioid usage [22]. Within our study, this finding was relevant to a postmyeloma patient reporting a 50% decrease of opioid dosages and a 30% reduction in his thoracic pain post-BVNA. Additionally, the use of RF ablation has been proven to be safe and effective for achieving pain reduction for treatment of VBP from vertebral body tumors. Koo et al. [26] retrospectively analysed patients with metastatic bone disease not suitable for excision, and found clinically significant VAS changes at 1-, 4-, 12-, and 24-week post-RF ablation. Another study found 15% of metastatic patients experienced radicular symptoms following the procedure, which ceased after transforaminal blocks [27]. The results found in our study are consistent with that of the current literature on RF ablation for this population.
Contraindications to BVNA proposed by Tieppo Francio et al. [16], to some extent limit the scope of expanded indications and include active systemic infections, pregnancy, skeletal immaturity, prior fusion at the treatment level, type 3 Modic changes, and presence of implantable pulse generators.
Potential complications of the procedure arise typically from inappropriate needle placement which may cause nerve root injury, radiculopathy, transient radiculitis, anterior vascular laceration or peridural hematoma [16]. Further complications related to pedicle access and heating may rarely lead to fracture or avascular necrosis, compared to general consequences of positioning and anaesthetic. The aforementioned technical complications highlight the importance of frequent fluoroscopy/CT view checks as the introducer is advanced [28]. There has been concerns regarding deep vertebral access to the BVN plexus which has relevance to the expanded indication of symptomatic osteoporotic and pathological fractures. The osseous access tools used in the Intracept procedure creates minimal damage to trabecular bone structure [29]. The RF energy heats the trabecular bone and causes transient weakening prior to subsequent healing. This process is time-dependent with progressive mineralisation over the first 4 weeks following the ablation [29]. Theoretical concerns extend beyond short-term pedicular or new vertebral fracture to osteonecrosis. The procedural costs associated with the Intracept system are recognised to be relatively high, which may influence patient satisfaction and uptake depending on financial accessibility. This is particularly relevant in settings where private health insurance or reimbursement pathways are limited. To combat this, a bipolar setup may be more economical and accessible to proceduralists in different socioeconomic settings.
Vertebral compression fractures post-BVNA is a device-related adverse effect which is infrequent but reported in research [12,24]. Findings suggest that BVNA has a transient negative impact on bone strength that predisposes high risk osteoporotic patients to a vertebral compression fracture. Conversely, previous phase I studies have shown no long-term adverse effects on the vertebral body after a period of mineralisation and healing [12]. Therefore, it is recommended that preexisting osteopenia or osteoporosis should be treated pharmacologically where feasible in advance of BVNA to enhance bone strength and reduce the risk of compression fractures [12]. Drawing from the current body of evidence, BVNA demonstrates a robust safety profile and favourable tolerability in appropriately selected patients, provided attention is given to patient selection criteria and procedural precision.
Per the constraints of our review, there are 40 patients with data collection from a single surgeon and musculoskeletal interventional radiologist. Although the expanded indications group comprised a smaller cohort (n=8) relative to the Modic type 1 or 2 group (n=32), their clinical outcomes were comparable, with no evidence of inferiority observed. This case series utilised the Likert scale as a subjective secondary outcome measure, relying on patient’s experience which augmented the possibility of bias albeit that this subjective measure was done at the 2-month appointment where the efficacy of the treatment would be expected to be representative. A limitation of this study is that formal DEXA-derived T- or z-scores were not uniformly available, and bone quality could only be assessed retrospectively via available osteoporosis scores or quantitative HU measurements of vertebral body cancellous bone. Further, the short-term follow-up of only 2 months is a limitation, however, randomized double‐blind sham‐controlled multi‐centre studies suggest efficacy even at 5-year follow-up [24] where most results are likely to be stable at midterm follow-up based upon this published experience. Contamination from staged or concurrent non-BVNA procedures may have influenced the subjective experiences of patients. This effect being significant is unlikely as, outlined by Tieppo Francio et al. [16], ODI and pain scores do not differ between groups of ‘BVNA only’ to ‘BVNA + procedure/intervention,’ which included additional treatment within the follow-up period either related to the presenting complaint or not. Additionally, patients who were satisfied may have been more likely to respond, thus biasing the results. Nevertheless, the high response rate and lack of clinical difference between groups 1 (t=0.539, df=38.0) and 2 (t=-0.915, df=38.0) suggest that this was not the case. In our cohort, all patients were available for follow-up and answered.
Further research is essential to validate and compare results of expanded with the well described indications. The sham-controlled RCT by Fischgrund et al. [24], established efficacy versus placebo for the indication of VBP due to DDD but not against expanded indications. Well-designed studies conducted in a controlled cohort fashion, ideally incorporating placebo control arms, are essential to further substantiate the role of BVNA in the management of spinal disorders beyond degenerative Modic type 1 and 2 MRI endplate signal changes. Expanding the evidence base to include indications such as tumor- or fracture-related vertebral pain, inflammatory endplate changes, and spinal deformities in patients undergoing pre-fusion evaluation has the potential to significantly enhance patient outcomes and overall quality of life in otherwise morbid populations who are often unsuitable for aggressive surgical management.
CONCLUSION
The early clinical experience with RF BVNA utilising the bilateral bipolar transpedicular technique highlights its potential as a minimally invasive intervention with reasonable efficacy and versatility in different socioeconomic settings. This technique demonstrates a robust safety profile and may play a significant role in managing patients with expanded indications, especially those who have failed conservative treatments and are unwilling or unsuitable for surgery, either presently or in the future. Furthermore, it offers a promising alternative to more invasive procedures such as fusion or disc replacement for VBP in appropriately selected cases.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation 
Print 
