Authors: Soun Sheen, Ryan S. D'Souza, Michael Stauff, Steven P. Cohen, Jay Karri
Categories: Editorial
Source: Interventional Pain Medicine
Authors: Soun Sheen, Ryan S. D'Souza, Michael Stauff, Steven P. Cohen, Jay Karri
Cervical radiofrequency ablation (RFA) is a well-supported, evidence-based intervention for treating facet-mediated neck pain. However, its efficacy may be reduced in postsurgical patients with altered cervical medial branch nerve (MBN) integrity. This review examines how common cervical spine surgical approaches affect MBN integrity and proposes a pragmatic framework for assessing RFA candidacy in postsurgical patients.
Anterior surgical approaches typically preserve the posterior elements and spare the MBNs, thereby maintaining RFA candidacy at operated and adjacent levels. In contrast, posterior cervical fusion commonly disrupts the MBNs at instrumented segments due to lateral mass exposure and decortication, making RFA at fused levels unlikely to be effective; however, adjacent levels may remain appropriate targets with careful technical planning. Laminoplasty carries a moderate-to-high risk of MBN compromise because of posterior element dissection and partial facet violation, requiring individualized evaluation using operative reports and postoperative imaging. Minimally invasive surgeries including laminotomy and foraminotomy more often preserve MBN integrity, particularly when facet resection is limited.
Successful cervical RFA in postsurgical patients requires a systematic pre-procedural evaluation – correlation of symptoms with facet-mediated pain patterns, review of surgical history and postoperative imaging to determine MBN integrity, and assessment of technical feasibility including hardware proximity and fluoroscopic access. While patients generally remain candidates after undergoing anterior cervical spine surgeries, posterior fusion patients require avoidance at instrumented levels, and non-fusion posterior procedures warrant individualized assessment. This anatomy-based framework may improve patient selection, though prospective validation is needed.
Determining candidacy for radiofrequency ablation (RFA) treatments in patients who have undergone cervical spine surgery remains a unique clinical challenge. This population includes both patients with new-onset facet-mediated neck pain after surgery, often attributable to adjacent segment disease, and those with persistent facet-mediated pain not completely addressed by the initial operative intervention. Even with technically successful cervical spine surgery, up to 40% of patients report continued axial neck pain, and its etiology is often multifactorial and biomechanically driven [1]. Fusion surgeries, such as anterior cervical discectomy and fusion (ACDF) and posterior cervical fusion (PCF), eliminate motion at instrumented segments, which may increase biomechanical stress at adjacent facet joints [2]. Whereas motion-preserving options aim to reduce this risk, all such surgical approaches – including cervical disc arthroplasty, posterior foraminotomy, laminoplasty, or laminectomy – alter spinal kinematics and may heighten the risk of at-level or adjacent-level facet joint pathology [3,4]. In decompression surgeries that preserve the facet joints, the decreased structural support may increase segmental motion and joint stress [5].
Cervical RFA is a well-established intervention for treating facet-mediated neck pain [[6], [7], [8], [9], [10]]. Its efficacy depends on several key factors – accurate identification of facet-mediated pain, anatomical and technical feasibility, and the integrity of the cervical medial branch nerves (MBN). Surgical dissection, decompression, or instrumentation can permanently disrupt these nerves, potentially rendering RFA ineffective at surgically altered segments [11]. Although based on low-quality evidence, this tenet is supported by studies conducted in the lumbar spine, but not in the cervical spine [1,12]. This creates a critical evaluation clinicians must determine which postsurgical patients remain viable candidates for RFA based on an understanding of how different surgical approaches impact cervical spine kinematics and MBN integrity.
Current guidelines acknowledge postsurgical patients as potential candidates for cervical RFA, yet they fail to provide the nuanced recommendations needed for clinical decision making in this distinct patient subgroup. This contrasts with the lumbar spine, where Medicare and other guidelines explicitly disqualify facet blocks at fused levels [13]. Cervical-specific guidelines do not provide a parallel clarification, likely reflecting the prevalence of anterior surgical approaches, which preserve posterior elements and therefore do not inherently preclude facet-based interventions. The available literature does not adequately stratify outcomes based on surgical approach or level, and nerve blocks of the MBNs in this population may also have higher false-positive rates, suggesting limited diagnostic accuracy in postsurgical patients [1]. Consequently, there is no clear framework for evaluating cervical MBN integrity and RFA candidacy relative to a patient's specific surgical history. This review addresses this critical gap by characterizing anatomical approaches utilized for common cervical spine surgeries and their potential impact on MBN integrity. We provide a practical, anatomically grounded framework to guide clinicians in evaluating RFA candidacy, based on MBN integrity and technical feasibility, for facet-mediated neck pain in patients who have undergone cervical spine surgery.
The C3-C4 through C7-T1 facet joints receive dual innervation from the MBNs of the cervical dorsal rami at the index level and the level above. For example, the C3-4 facet joint is innervated by the C3 and C4 MBNs. An important exception is the C2-C3 facet joint, which is innervated by the third occipital nerve (TON) and occasionally by the C3 deep MBN [3,14]. At C4 and C5, >15% of individuals may have dual medial branches, though given their close proximity, the anatomical significance of this for treatment is unknown [15]. From their origin at the dorsal rami, the cervical MBNs course posteriorly along the waist of the articular pillars (Fig. 1). Whereas their general pathway is typically consistent in the sagittal plane (along the ventral-dorsal axis), key variations exist in their course along the articular pillars in the coronal plane (superior-inferior axis). A plexus-like innervation has also been described in the upper cervical spine, wherein articular branches may arise directly from the dorsal root [16]. After passing the articular pillars, the MBNs extend further posteriorly to innervate the deep cervical musculature. They run between the semispinalis capitis and cervicis before their terminal branches arborize within the multifidus [14].Fig. 1Axial view of normal cervical anatomy demonstrating the course of the medial branch nerve as it exits the dorsal ramus, crosses laterally over the waist of the articular pillar.
A comprehensive understanding of this anatomy is paramount for assessing candidacy and technical planning for cervical RFA procedures, especially in patients with prior cervical spine surgery. Such assessments are complicated by the inherent anatomical variability of the cervical MBNs and the structural alterations caused by surgical interventions. Specifically, surgeries involving posterior dissection, hardware placement, or placement of bone graft material risk inadvertent injury to or obscuration of the MBNs. Therefore, integrating knowledge of normative anatomy with patient-specific surgical changes is essential for appropriate patient selection for cervical RFA procedures.
Anterior surgical exposures are performed via the Smith-Robinson approach to access the anterior vertebral bodies, disc spaces, and the anterior cervical spinal canal and foramina. The anterior approach to the cervical spine allows access for the performance of various procedures that include ACDF, cervical disc arthroplasty, anterior corpectomy, and anterior endoscopic discectomy. The standard dissection for anterior access proceeds through a fascial plane deep to the platysma, medial to the sternocleidomastoid and carotid sheath, and into the prevertebral space, entirely sparing the posteriorly positioned articular pillars and MBNs [17] (Fig. 2).Fig. 2Illustration of anterior access. The MBN preserved during anterior surgical access is shown as a blue square. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
Clinical Implications:●Patients with prior anterior cervical spine surgery typically have intact MBNs, and RFA remains anatomically feasible at both fused and adjacent segments; however, clinical evidence supporting its efficacy in this population is limited.●Facet-mediated pain in this population may originate from adjacent segments due to altered biomechanical loading and, less commonly, from at-level facet joints secondary to posterior element overload.
In contrast, a posterior surgical approach requires dissection through or near the posterior elements, introducing a variable risk of injury to the MBNs (Fig. 3). This risk ostensibly correlates with the extent of lateral exposure and facet joint manipulation. The surgery typically begins with field exposure through the nuchal ligament and paraspinal musculature in a subperiosteal manner out to the lateral aspect of the lateral masses at all operative levels. The surgery can involve a cervical laminectomy, which involves removing the spinous processes and laminae across one or more levels to decompress the spinal canal. This decompression is commonly accompanied by a posterior cervical fusion. The instrumentation for a posterior fusion involves either pedicle screw or lateral mass screw placement. The less common pedicle screw technique requires a more medialized trajectory for screw placement, which might necessitate additional lateral dissection and exposure to identify landmarks and safely place implants (Fig. 4A). Due to this extensive lateral dissection, the cervical MBNs might be sacrificed at instrumented levels. More commonly, surgeons use lateral mass screws for fixation in the subaxial cervical spine. Lateral mass screw placement requires exposure of the entire posterior lateral mass, but may not require the same degree of lateral exposure as the pedicle screw trajectory (Fig. 4B). The surgical approaches for posterior cervical decompression and fusion are likely to disrupt MBN integrity at instrumented levels, thereby making RFA at those segments anatomically unreliable. Whether facet joints at successfully fused levels remain clinically relevant pain generators is uncertain, given the current lack of preclinical and clinical evidence. Posterior instrumentation may also impair fluoroscopic visualization or limit safe RFA probe placement, rendering RFA technically challenging in this subgroup.Fig. 3Illustration of posterior access, highlighting the likely disruption of the posterior elements, including the MBN, shown as red square. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)Fig. 4A. Illustration of pedicle screw placement, which often requires additional lateral dissection. B. Illustration of lateral mass screw placement, which often does not require the same degree of lateral dissection. The MBN is shown as a red square. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
Clinical Implications:●RFA is likely ineffective at levels fused posteriorly, given the high likelihood of MBN disruption. Posterior instrumentation may also pose additional procedural challenges due to limited fluoroscopic visualization or potential thermal conduction injury, raising additional concerns as to whether RFA can be performed safely in this subgroup.●RFA at adjacent, non-fused levels might be considered but requires careful technical probe placement to mitigate risk for thermal conduction near surgical hardware.
Cervical laminoplasty is a motion-preserving surgery that decompresses the spinal canal without a fusion. While there are different techniques for cervical laminoplasty, they all require a longitudinal posterior incision and subperiosteal dissection of the cervical paraspinal muscles to expose the lamina, spinous process, and lateral masses [18,19]. “Troughs” are then created at the lamino-lateral mass junction, typically across multiple levels, and subsequently stabilized with plates secured to the lamina and lateral mass at the same level (Fig. 5A). The most common laminoplasty techniques include the open-door and French-door techniques. Generally, both techniques involve exposure of the posterior cervical spine extending out to the lateral masses bilaterally, but the required exposure is not as wide as for posterior cervical fusion. The approach for laminoplasty could pose a plausible risk of ipsilateral cervical MBN or facet joint injury [20]. Some studies have found a high incidence of facet joint violation with certain approaches [21]. Careful review of operative reports and postoperative imaging is essential to identify possible cervical MBN or facet joint compromise and thus optimize patient selection for cervical RFA.Fig. 5A. Illustration of cervical laminoplasty demonstrating the open-side hinge creation, which often requires exposure lateral to the facet joint. The MBN is shown as a red square. B. Illustration of cervical foraminotomy demonstrating reduced lateral exposure. The MBN is shown as a blue square. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
Clinical Implications:●Patient selection for RFA requires an individual assessment of the operative reports and postoperative imaging to identify the precise location of hardware placement and potential MBN disruption. For example, patients who have undergone laminoplasty may have greater MBN integrity and be more likely to benefit from cervical RFA than those who have gone posterior instrumented fusion.
Posterior laminotomy and foraminotomy procedures also require a posterior midline or paramedian incision with dissection of the cervical paraspinal muscles down to the lamina. The surgical trauma is often less extensive than that required for PCF or laminoplasty, both of which require more lateral exposure for hardware placement. Therefore, the articular pillar and facet joints are often preserved unless additional facetectomies are performed [22] (Fig. 5B). However, concomitant facetectomies performed with foraminotomy procedures are often only partial in nature and medial in location, thus preserving a majority of the joint complex for stability and as potential pain generators [23]. Therefore, the facet joints and articular pillars are largely preserved. In vitro studies have shown increased stress on the (mostly ipsilateral) posterior elements, including the facet joints, after posterior laminotomy and/or foraminotomy [24].
Endoscopic techniques may further enhance preservation of the facet joints and the MBNs by minimizing collateral soft-tissue disruption and limiting facet joint removal compared to open or microscopic approaches. Endoscopic posterior foraminotomy is associated with a significantly lower proportion of cases in which more than 50% of the facet joint is resected, supporting the role of endoscopic approaches in maintaining facet integrity and protecting the MBNs [25].
Clinical Implications:●Patients with laminotomy and foraminotomy surgeries typically have intact MBNs, and RFA remains a viable intervention.●In patients where additional facetectomies are performed, assessment of the operative reports and postoperative imaging is necessary to identify the extent of facet resection and hence the likelihood of its nociceptive potential, and the potential for MBN disruption.
A systematic pre-procedural evaluation is essential to assess cervical MBN integrity in postsurgical patients. It is best conceptualized as a dual approach addressing two distinct questions. The first question is should RFA be performed, a determination made by correlating symptoms with facet-mediated pain through history, physical examination, and correlation with established cervical facet referral patterns, as suggested per consensus guidelines on cervical facet-mediated pain [3]. The second question is can RFA be performed, determined by anatomical and technical feasibility based on surgical history and imaging to estimate anatomical preservation of these nerves and RFA candidacy.
The evaluation should begin with a detailed review of the operative report to determine the surgical approach, extent of lateral exposure, and any facet joint disruption. This information provides initial insights into whether the anatomical course of the MBNs is preserved. However, operative reports may not always be available; therefore, postoperative imaging should always be reviewed to corroborate these findings. Whereas plain radiographs can identify hardware placement, advanced imaging - namely computed tomography (e.g. CT scans)- can provide detailed visualization of the bony anatomy including the facet joints and the articular pillars. Together, these imaging modalities enable a segmental assessment of MBN integrity.
The likelihood of successful cervical RFA in postsurgical patients is fundamentally determined by the surgical approach (Table 1). Anterior approaches generally spare the MBNs, whereas posterior fusions generally disrupt them at instrumented levels. RFA at adjacent non-fused segments remains possible, but requires careful planning to avoid thermal conduction near spinal hardware [3,26]. Posterior decompression surgeries without fusions (e.g. laminoplasty, foraminotomy) require individualized assessments, as the degree of lateral dissection and facet involvement varies widely.Table 1Summary of facet joint and MBN integrity by cervical surgical approach.Table 1SurgeryFacet Joint IntegrityExpected MBN IntegrityClinical Considerations for RFAAnterior surgical approachesACDFCervical Disc Arthroplasty Corpectomy Endoscopic DiscectomyPreservedIntactProceed if indicated.Facet-mediated pain may be more prevalent at adjacent segments. RFA can be performed without high concern for thermal conduction to hardware.Posterior surgical approachesPosterior decompression with fusionCompromised at instrumented levelsCompromised at instrumented levelsAvoid at fused levels. Facet-mediated pain may be less common in this population.Adjacent non-fused levels may be considered with caution. Review imaging to confirm hardware and plan probe placement away from hardware to minimize thermal injury risk.LaminoplastyTypically preservedModerate-High risk of compromiseIndividualized assessment required. Higher likelihood of RFA benefit when performed with central plates or spacers, minimizing lateral mass/facet exposure. Review operative notes and imaging.LaminotomyForaminotomyTypically to mostly preservedLow to Moderate risk of compromiseIndividualized assessment required. Higher likelihood of RFA benefit when facet resection is <50% and posterior elements are intact on imaging. Surgery may increase stress on the facet joints, making it more likely they are pain generators. Review operative notes and imaging.
Applying standard RFA protocols indiscriminately risks treatment failure in patients with prior cervical spine surgery. Medial branch nerve blocks may offer prognostic value in selected cases, though false-positive rates are higher post-surgery [12], necessitating consideration of controlled blocks to enhance the positive predictive value. Nerve blocks should target levels with the highest likelihood of preserved anatomy, and responses must be interpreted in context of surgical history and imaging. In the cervical spine, the medial branches innervate not only the multifidus muscle but also the semispinalis cervicis and capitis. Hence, motor stimulation of these nerves during radiofrequency procedure often elicits profound paraspinal muscle contractions in individuals with preserved neuroanatomy. However, before the facet denervation stage, there are no validated electrophysiological testing protocols that can confirm postoperative MBN integrity. This uncertainty, compounded by high surgical variability, necessitates a tailored, anatomy-driven approach rather than standardized protocols. Prospective studies correlating anatomical predictions of MBN disruption with RFA outcomes are needed to validate this framework.
RFA of the cervical MBNs is a potentially effective treatment for facet-mediated neck pain in patients with prior spinal surgery. However, its success requires a comprehensive pre-procedural framework that integrates the following key elements – appropriate patient selection based on clinical presentation and physical examination, assessment of MBN integrity with surgical history and imaging, and technical accessibility of the target nerves. Anterior surgical approaches generally spare the MBNs, preserving the anatomical substrate for RFA in clinically appropriate candidates. However, clinical evidence supporting its efficacy specifically at anteriorly fused segments is limited. Therefore, anatomical feasibility alone should not be interpreted as an established indication. In contrast, posterior fusion procedures typically disrupt MBNs, thereby contraindicating RFA at fused segments; however, adjacent levels may still be appropriate procedural targets. Posterior non-fusion surgeries require individualized evaluation of facet integrity via imaging and operative reports. Applying a generic RFA protocol risks both treatment failure and unnecessary procedures; therefore, a tailored and anatomy-based approach is essential. Well-designed prospective studies correlating pre-procedural anatomical predictions with the above framework and clinical outcomes are needed to validate this individualized paradigm.
Not applicable, no direct patient data were analyzed or reported.
JK and SS conceived study idea and authored the manuscript. RSD, MS, and SC assisted with manuscript review and edits.
Not applicable.
The authors have no sources of funding to declare for this manuscript.
There are no conflicts of interest to declare for this manuscript.