INTRODUCTION
Chronic low back pain (CLBP) is a prevalent and debilitating condition affecting a significant portion of the global adult population. Defined as pain lasting 12 weeks or more, CLBP imposes substantial healthcare costs and severely impacts the individuals’ quality of life [1]. A 2019 global study ranked low back pain (LBP) among the top 10 contributors to the Global Burden of Disease, positioning it on par with heart disease, diabetes, and cancer in terms of its global health impact [2]. With a prevalence of 3% to 10%, CLBP is associated with complications such as depression, immobility, and loss of work capacity, thus placing a profound impact on both individuals and healthcare system [3]. Incidence peaks at 27% among individuals in their fifth decade of life [4].
The etiology of CLBP are complex and multifactorial, encompassing various anatomical components, including lumbar intervertebral discs, sacroiliac joints, dura mater, fascial tissues, ligaments, and muscles. These structures contribute differently to the manifestation of pain [5]. In particular, the lumbar facet joints (LFJs), synovial joints richly innervated by the dorsal medial branch (DMB), are susceptible to pathologies like inflammation and degeneration, which can lead to facet joint syndrome (FJS). FJS, characterized by localized pain in the lumbar spine, accounts for 15% to 45% of CLBP cases [6,7]. The LFJ’s exclusive sensory innervation by the DMB makes medial branch denervation a critical approach for treating facetogenic CLBP.
Various treatment modalities are available for facet joint (FJ)-related CLBP, including fluoroscopy-guided medial branch blocks (MBBs) combined with intra-articular steroid and anesthetic injections. These interventions aim to temporarily disrupt pain transmission from the affected facet joints (FJs). Another prominent treatment is medial branch radiofrequency (RF) neurotomy, commonly referred to as rhizotomy, which has been shown to provide sustained pain relief based on sound anatomical principles [8-11]. Traditional percutaneous RF ablation (RFA) is a minimally invasive technique but lacks direct visualization of the DMB, limiting its precision [12].
In contrast, endoscopic rhizotomy (ER) has emerged as a feasible alternative for managing CLBP. This technique allows for direct visualization of the DMB and adjacent spinal structures during denervation, enhancing precision in severing the small branches of the DMB and identifying anatomical variations [13]. This study reviews the efficacy of ER in the treatment of facetogenic CLBP and compares clinical outcomes between ER and RFA patient cohorts, offering a comprehensive overview of this evolving therapeutic option.
METHODS
1. Materials and Methods
This systematic review was carried out in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines [14]. A comprehensive systematic search was conducted across the Embase, PubMed, and Cochrane Central Register of Controlled Trials online databases, covering the literature from each database’s inception until February 1, 2024. The search strategy employed a predefined set of keywords, including: low back pain, facetogenic low back pain, RFA, full-endoscopic lumbar rhizotomy and comparative study. Additionally, reference lists from relevant articles were thoroughly reviewed to identify any further studies that met the inclusion criteria (Figure 1).
2. Study Selection
The identification and exclusion of relevant studies were conducted independently by 2 authors (CJC and YFL). Discrepancies between the reviewers were resolved through consensus discussions to ensure adherence to the predefined inclusion and exclusion criteria. Inclusion criteria consisted of studies involving patient groups undergoing full-endoscopic lumbar rhizotomy, as well as comparative studies evaluating ER versus conservative treatments or traditional percutaneous radiofrequency ablation (TPRA). Eligible patients were those with CLBP lasting for more than 12 months. Additional inclusion criteria required studies to achieve a Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) score of 16 or higher [15]. To ensure the inclusion of studies that provided at least short-term outcomes, a minimum follow-up period of 2 months was selected, as this duration is commonly used to capture early postoperative results. Exclusion criteria were stringently applied to omit non-English studies, unpublished manuscripts, studies with incomplete patient demographic information, and those lacking sufficient data. Following these criteria, selected studies underwent a detailed evaluation. Each study’s quality was meticulously assessed using STROBE guidelines to ensure a systematic and consistent review process.
3. Data Extraction
The extracted data from the included articles encompassed the following key elements: (1) Patient characteristics, which comprised age, gender, duration of back pain, postprocedure follow-up period, and the specific rhizotomy approach used. (2) Clinical outcomes, including the visual analogue scale (VAS), MacNab scale, global impression of change (GloC), EuroQoL-5 dimensions (EQ-5D), Oswestry Disability Index (ODI), and any documented complications. This systematic approach to data extraction ensured a comprehensive and standardized method for collecting information from the selected studies, thus enhancing the reliability and consistency of the findings.
4. Statistical Analysis
Outcomes analysis was conducted using the variance-weighted means. The degree of heterogeneity across studies was quantified using the I2 statistic, which ranges from 0% to 100%, with values above 50% indicating substantial heterogeneity. In cases of significant heterogeneity, a random-effects model was applied for data analysis, while a fixed-effects model was used for studies with minimal heterogeneity. This approach to assessing heterogeneity and calculating mean differences was systematically applied across all outcomes in the analysis. Additionally, potential publication bias was evaluated using Egger's funnel plots. The analysis was performed using Review Manager (RevMan 5.3, Cochrane Collaboration, Oxford, UK) provided by The Cochrane Collaboration, which facilitated data synthesis and the graphical presentation of results.
RESULTS
After applying the predefined inclusion and exclusion criteria, a total of 10 studies were deemed suitable for inclusion in the systematic review. The patient characteristics across 10 these studies are summarized in Table 1 [16-25]. Together, these studies involved a cumulative cohort of 529 patients. The duration of recorded back pain ranged from 24 months (midterm) to 137 months (long-term). In most studies, the risk of bias was assessed as unclear or low. A summary of the risk of bias assessment is presented in Figure 2.
Three of the 10 included studies reported MacNab score [17,23,25], with all indicating that 96.7% to 97.8% of treated patients achieved excellent or good postoperative outcomes. Regarding clinical outcomes, Walter et al. reported significant improvements in subjective well-being at the final follow-up compared to pretreatment levels, as measured by the core outcome measures index and EQ-5D [22]. Similarly, Du et al. [24] found that, 12 months posttreatment, GloC scores demonstrated significantly greater improvements in the ER group compared to the TPRA group (Table 2).
In the pooled data analysis, patients who underwent ER showed significantly better VAS scores compared to those in the TPRA group (mean difference: -2.39; 95% CI, -3.15 to -1.63; p<0.00001, random-effects model). The heterogeneity for this outcome was moderate (I2 = 79%), which supports the use of a random-effects model to account for variability among studies. Additionally, the ER group demonstrated superior performance on the ODI (mean difference: -19.97; 95% CI, -37.48 to -2.46; p=0.03, random-effects model). However, substantial heterogeneity was observed for the ODI outcome (I2=98%), indicating greater variability across the included studies, which was addressed using a random-effects model (Figure 3). Notably, no major postoperative complications were reported in any of the included studies (Table 3).
DISCUSSION
The pooled data from our analysis showed that patients with CLBP who underwent ER reported superior pain relief, as reflected by better VAS scores compared to those treated with TPRA. Furthermore, the ER group exhibited significantly greater improvements in functional outcomes, as measured by the ODI. The effectiveness of ER is further evidenced by the remarkable absence of major side effects or complications in the studied patient populations.
The diagnosis of facetogenic pain remains a significant challenge due to the high prevalence of CLBP and the relatively low accuracy of diagnosing FJ pain specifically. This complexity is compounded by the need for precise diagnostic procedures to confirm the FJ as the source of pain [26]. In clinical practice, patients often present with characteristic symptoms, such as axial back pain that improves when lying down and worsens with certain activities, which may suggest FJ involvement. These clinical signs are crucial for guiding further diagnostic and therapeutic interventions [27].
The anatomical distribution of the lumbar joint's synovial membrane and capsule, which are rich in nerve endings primarily originating from the medial branch of the dorsal ramus of the spinal nerve, plays a critical role in pain transmission. The intricate anatomy of these nerve branches underscores the challenge of accurately targeting the FJ for effective pain relief [28]. A key diagnostic tool, positive infiltration- where significant pain relief is observed within 3 hours of the procedure-is a critical before proceeding with RFA of FJ. This diagnostic approach requires patients to exhibit symptomatic pain relief following periarticular infiltration (specifically targeting the medial branch of the dorsal ramus), while ruling out other pathologies such as vertebral body fractures or malignancies through clinical and radiological evaluation [12,29]. Research by Boswell et al. [10] further supports that patients who experience pain reduction after infiltration tend to have higher success rates while RFA targeting the DMB. This evidence underscores the importance of accurate diagnosis and targeted treatment in improving outcomes for patients with facetogenic CLBP.
The management of CLBP typically begins with a broad range of conservative treatments; including nonsteroidal anti-inflammatory drugs, glucocorticoids, and opioids, aimed at providing symptomatic relief. These medications are chosen for their ability to reduce inflammation, alleviate pain, and minimize discomfort, offering a noninvasive approach in the initial stages of CLBP treatment [30,31]. However, the efficacy of these treatments is often limited by their temporary effects and the risk of adverse reactions, prompting the need for more targeted therapeutic strategies. For FJS, standard treatment modalities include intra-articular anesthetic steroid injections, MBBs, and RFA of the DMB. In recent years, percutaneous denervation of the FJ has emerged as a pivotal treatment option, often considered the "gold standard" for managing facetogenic pain [32]. The significance of this method is supported by a recent meta-analysis which highlights the efficacy of RFA in reducing low back pain associated with FJ inflammation for up to 12 months. Success in RFA treatment is closely linked to prior diagnostic blocks with local anesthetics, underscoring the importance of accurate preliminary diagnosis for favorable outcomes [33]. Nevertheless, anatomical variations in the medial branches, along with the presence of scar tissue in patients with prior spinal surgery, pose significant challenges to achieving optimal outcomes with these treatments. The heterogeneity of the nerve branches can impede precise targeting, a critical factor for successful denervation in the management of facetogenic pain. This challenge is particularly pronounced when direct visualization of the target area is obstructed, necessitating meticulous planning and precision during the execution of these techniques [28,34]. Furthermore, the recurrence of back pain can occur due to the potential regeneration of the medial branch of the dorsal ramus [16].
The term 'facet syndrome' was first introduced by Ghormley in 1933, identifying the FJ as a potential source of referred pain and sciatica due to direct root compression [35]. This concept was further expanded by Badgley, who explored the FJ's role in generating referred pain, sparking extensive research into its clinical significance, diagnostic methods, and treatment options [36-38]. The first interventional rhizolysis for CLBP was performed by Dr. Vincent Nesfield in 1918, who pioneered a technique involving the insertion of an ophthalmic scalpel into the spinal region to sever the entrapped nerve [39].
A major advancement in CLBP treatment occurred in 1975 the introduction of RF denervation by Shealy. His study, which monitored 207 patients over a period ranging from 6 to 21 months, reported pain relief in 79% of patients who had not undergone surgery and in 27% of patients with a prior spinal fusion [40]. This method established RF denervation as a reliable and effective intervention for managing CLBP [41]. In 1976, Mooney and Robertson [42] introduced radiographically guided injections of steroids and local anesthetic directly into the FJ, significantly improving diagnostic precision and therapeutic outcomes. Another pivotal contribution came from Bogduk in 1982, whose research on the clinical anatomy of the cervical dorsal rami advanced spinal surgery techniques, particularly the concept of ER. He highlighted the importance of the lumbar mamillo-accessory ligament (MAL), which connects the mamillary and accessory processes of lumbar vertebrae, forming an osseofibrous tunnel that houses the medial branch of the dorsal ramus. This discovery improved the precision and effectiveness of ER, offering an advantage over traditional RFA methods that lack direct visualization. However, Bogduk [43] noted challenges, such as the ossification of the MAL in 10% of lower lumbar vertebrae, which could complicate percutaneous denervation techniques.
Traditional percutaneous RFA, while minimally invasive, lacks direct visualization of the DMB, potentially affecting procedure precision. Furthermore, the durability of pain relief following RFA has been questioned, as a significant proportion of patients experiencing pain recurrence within 1 year, suggesting a relatively short-lived benefit of the treatment [23]. The efficacy of RFA, particularly for sacroiliac joint pain, tends to diminish over time due to nerve regeneration postlesioning. This concern is supported by findings from Chen, who reported a gradual return of pain and functional disability within 6 to 12 months after treatment, highlighting the need for more than temporary ablation to achieve long-term relief [25]. The high recurrence rate of pain within 12 months of traditional medial branch RF treatment often necessitates repeated interventions. This has led to the development of more advanced techniques, such as cooled RF, the 2-needle RF technique, and parallel electrode approaches, which aim to enhance nerve coagulation [44,45]. Despite these innovations, the median duration of pain relief following RF treatment remains around 9 months [46], with systematic reviews indicating that symptom relief typically wanes between 7- and 9-month posttreatment [47].
Anatomical variations in the medial branch of the dorsal ramus significantly impact the precision required for successful neurotomy. Incomplete ablation of the targeted nerve tissue, incorrect electrode placement, and the potential for nerve regeneration post-ablation are key factors contributing to the reduced efficacy of RFA procedures [48]. Chen’s study highlights the gradual recurrence of pain and functional disability following RFA, emphasizing the need to consider rhizotomy—a more definitive severance of nerve pathways—over simple ablation to achieve long-term resolution of sacroiliac joint pain [25].
The efficacy of RFA is further compromised by its limited heat distribution, which forms an oblate spheroid shape with a maximum width of only 2 mm along the electrode's axis. This narrow range of heat application often fails to fully encompass the target nerve tissue, resulting in incomplete nerve damage [49]. Moreover, the percutaneous puncture technique utilized in RFA is executed without direct visual confirmation, essentially making the procedure "blind." Surgeons rely on x-ray imaging to guide the electrode’s position surrounding bony landmarks, which increases the risk of inaccurate electrode placement and reduces the likelihood of effectively targeting the exact nerve location [20].
Traditional RFA methods are limited to ablating the target nerve at a single point, which may miss key areas contributing to pain. In contrast, endoscopic-guided RFA allows practitioner to trace the course of the nerve, interrupting it at multiple points using an endoscope and RF bipolar probe. This approach significantly increases the likelihood of achieving comprehensive denervation [24]. Moreover, the endoscopic technique enhances the visibility of spinal anatomy, including the intervertebral foramen, transverse processes, superior articular processes, and MALs, facilitating the accurate identification of treatment areas.
Advancements in endoscopic technology enables the detailed observation of sensory nerve fibers, even those as thin as 0.21 to 1.51 mL in diameter, improving precision during procedures [50]. Patient selection for ER or RFA, especially in individuals with a history of pedicle screw fixation, is influenced by unique challenges. Du’s study concluded that fluoroscopic imaging used to guide the RF cannula may be compromised by metal screws, as the RF cannula tip may overlap with the hardware in the images, complicating the procedure [24]. Cadaver studies have also demonstrated the potential for heat transmission from the RF cannula to the pedicle screws, which can cause unintended thermal injury to surrounding tissues [51].
Surgical indications for endoscopic FJ rhizotomy include CLBP that persists despite conventional treatments, FJ hypertrophy, and FJ arthritis-conditions characterized by degenerative changes that lead to significant pain and functional impairment. Post discectomy syndrome, defined by persistent pain following disc surgery, is another condition treatable by ER. However, several contraindications must be carefully considered to ensure patient safety and procedural effectiveness. Patients on anticoagulation therapy are generally advised against this procedure due to the increased risk of bleeding. Negative preoperative infiltration testing, which suggests that FJ pain is not the primary source of discomfort, also serves as a contraindication. The use of a monopolar RF probe is contraindicated in patients with a pacemaker, as it can interfere with device function. Finally, the presence of other pathologies, such as tumors at the affected spine level, may also preclude the use of ER, necessitating alternative treatments to effectively address the patient's condition [12].
In terms of surgical complications, Xue et al. [23] reported that the incidence of complications in patients undergoing ER was significantly lower compared to those in the control group. Additionally, other research included in our review demonstrated an absence of major complications following ER procedures (Table 3).
LIMITATIONS
This study's findings are subject to several limitations that should be acknowledged. First, the selection of only 10 studies, along with the limited availability of comparative research that met our selection criteria, raises the potential risk for publication bias. Moreover, the duration of follow-up varied across the studies, ranging from short to midterm (2 to 24 months), which may affect the evaluation of long-term functional outcomes and the identification of delayed complications. Furthermore, inconsistencies in outcome reporting and incomplete baseline data for some patients may introduce bias when pooling the results. Given these limitations, caution is advised in interpreting the outcomes of this systematic review. There is a pressing need for larger, high-quality randomized controlled trials to validate these preliminary findings.
CONCLUSIONS
In conclusion, our findings suggest that ER provides superior postoperative outcomes compared to TPRA. ER offers a viable and effective treatment option for patients with CLBP over a midterm period of 24 months to a long-term period of 137 months. Importantly, these benefits were achievable without the occurrence of major complications. These findings should be interpreted with caution due to the small number of included studies, variability in follow-up durations, and the potential risk of publication bias. Further high-quality randomized controlled trials are needed to validate these preliminary results.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation 
Print 
