Percutaneous endoscopic lumbar discectomy combined with sinuvertebral nerve ablation for lumbar disc herniation with Modic type I endplate changes: a retrospective case-control study

Article information

Asian Spine J. 2026;.asj.2025.0752

Publication date (electronic) : 2026 July 8

doi : https://doi.org/10.31616/asj.2025.0752

Lei Liu, Meng Han

Department of Spine Surgery, Xuzhou Central Hospital, Xuzhou, China

Corresponding author: Meng Han, Department of Spine Surgery, Xuzhou Central Hospital, No. 199 Jiefang South Road, Quanshan District, Xuzhou, Jiangsu 221009, China, Tel: +86-15351687838, Fax: +86-051683840486, E-mail: hanmeng2010china@163.com

Received 2025 November 25; Revised 2026 January 26; Accepted 2026 February 11.

Abstract

Introduction

Lumbar disc herniation (LDH) is one of the most common spinal conditions that causes the clinical symptom of low back pain (LBP). Aberrant pain signal conduction by the sinuvertebral nerve and sensitization to pain stimuli further exacerbate LBP [1].

Modic changes (MCs) are associated with pain episodes of higher frequency and longer duration, as well as poorer outcomes following conservative treatment [2]. Patients tended to exhibit higher Visual Analog Scale (VAS) scores for LBP as the follow-up time increased, particularly those with LDH accompanied by Modic type I changes (MC I) [3]. MCs indicate severe endplate alterations and lumbar instability in patients, warranting consideration of fusion surgery [4]. Patients with MCs still experience LBP postoperatively [5], which may be alleviated by targeted treatment. Previous studies have shown that innervation of the endplate primarily arises from the sinuvertebral nerve [6,7]. Lumbar pain in patients with LDH can be alleviated through sinuvertebral nerve ablation (SNA) [8,9].

Percutaneous endoscopic lumbar discectomy (PELD) is widely used for the treatment of LDH. A study reported that PELD is significantly effective in treating LDH, regardless of whether it is accompanied by MCs [10]. However, postoperative back pain and the functional status of patients with MCs tend to deteriorate with time, especially in those with MC I. Persistent inflammatory reactions stimulated by inflammatory mediators acting on the sinuvertebral nerves located on the surface of the lumbar disc may cause postoperative LBP. The combination of PELD and SNA may enhance clinical efficacy, fully alleviate postoperative LBP, and provide a more comprehensive approach for the treatment of LDH [11]. Therefore, the present retrospective study aimed to evaluate the clinical efficacy of PELD combined with SNA based on the clinical data of patients with LDH accompanied by MC I.

Materials and Methods

This was a single-center retrospective case-control study. Patients with LDH admitted to Xuzhou Central Hospital between January 2022 and December 2022 were enrolled. Patients were divided into two groups based on the surgical method: patients who underwent PELD surgery combined with SNA were assigned to the PSNA group, and patients who underwent PELD only were assigned to the PELD group. This study was approved by the Ethics Committee of Xuzhou Central Hospital (No. XZXY-LK-20240305-0038). Owing to the retrospective nature of the study, the Xuzhou Central Hospital Institutional Review Board waived the need to obtain informed consent. All experiments were performed in accordance with relevant guidelines and regulations. This study consistently adhered to the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines for observational research.

The inclusion criteria were: (1) LDH diagnosis with PELD surgery after >6 months of ineffective conservative treatment; (2) LDH diagnosis based on patient symptoms and physical examinations, with confirmation by lumbar spine radiography, two-dimensional computed tomography (CT), and magnetic resonance imaging (MRI); (3) vertebral endplates of the operated segments exhibiting MC I; (4) age >18 years; and (5) follow-up time ≥12 months.

The exclusion criteria were: (1) surgery for LDH recurrence; (2) bilateral PELD or surgery more than one surgical level; (3) imaging (CT three-dimensional reconstruction and MR plain scan) suggesting infection, tumor, lateral recess stenosis (based on the width of the lateral sulcus on CT images), intervertebral foramen stenosis, lumbar spine deformity, or lumbar spondylolisthesis; (4) underlying disease affecting the observation of clinical symptoms; and (5) inability to review or follow up the patient regularly.

Surgical methods

PELD surgery

All procedures were performed by the same group of surgeons. A C-arm was used to locate the affected segment. Anesthesia was administered via local infiltration, one layer at a time, into the vertebral plate and articular synovial joints. A 1 cm incision was made, and a soft tissue expander was inserted along the guidewire, which was placed on the inner edge of the articular eminence under fluoroscopy. A working trocar was advanced along the expander, and a spinal endoscope was inserted through the working trocar to debride the soft tissue within the safety triangle of the intervertebral foramen. Partial resection of the vertebral plate, the bone of the articular synchondrosis joints, and the hypertrophied ligamentum flavum was performed. The dural sac and nerve root were exposed, revealing a herniated intervertebral disc compressing the nerve root. Radiofrequency ablation electrodes (VAM-Z-3040340; Shenzhen WeiAo Medical Technology Co. Ltd., Shenzhen, China) were used to secure hemostasis, dissect adhesions, and mobilize the nerve root and dural sac. Nucleus pulposus forceps were used to extract herniated disc tissues and loosen nucleus pulposus tissues within the intervertebral disc. The tip of the radiofrequency electrode was inserted deep into the disc for intradiscal nucleus pulposus molding. It was then withdrawn to the fibrous ring breach, and fibrous ring plication was performed. The dural sac regained mobility, and the nerve root appeared flaccid and retracted naturally. No active bleeding was observed, and the wound was sutured and dressed under aseptic conditions (Fig. 1A–C).

Fig. 1

Female, a 41-year-old, presented to our hospital due to low back pain for 1 year. Preoperative magnetic resonance imaging revealed L4/5 disc herniation, with Modic type I changes (MC I) at L4 and L5 endplates (A), accompanied by compression of the right nerve root (B). Based on physical examination, the patient was diagnosed with lumbar disc herniation. The patient underwent percutaneous endoscopic lumbar discectomy under local anesthesia. Panel (C) shows the nerve roots (black arrow), intervertebral discs (yellow arrow), and sutured fibrous rings (blue arrow) visualized under an intervertebral foramen endoscope.

SNA

The same methods as described above were used. Before the end of surgery, radiofrequency treatment was performed using radiofrequency ablation electrodes within the preoperatively designed radiofrequency ablation range. The radiofrequency ablation range was centered on the broken annulus fibrosus, with the inner edge of the intervertebral foramen on the outer side, midline of the spinal canal on the inner side, upper boundary 5 mm above the lower edge of the endplate of the upper vertebral body, and lower boundary 5 mm above the upper edge of the endplate of the lower vertebral body. The following tissues underwent radiofrequency ablation: posterior longitudinal ligaments, outer layer of the annulus fibrosus, localized proliferative vascular tissues, and connective tissues (Fig. 2A–C).

Fig. 2

Female, a 49-year-old, was admitted to the hospital due to lumbosacral pain for 2 years. Preoperative magnetic resonance imaging revealed L5/S1 disc herniation, with Modic changes in the L5 and S1 endplates (A), accompanied by compression of the right nerve root (B). Based on the physical examination, the patient was diagnosed with lumbar disc herniation. The patient underwent PSNA under local anesthesia. Panel (C) shows the nerve root (black arrow), intervertebral disc (yellow arrow), ruptured annulus fibrosus (blue arrow), and ablation of the sinuvertebral nerve (white arrow) within the target area using a high-frequency ablation electrode under endoscopic visualization. PSNA group, PELD+SNA; PELD, percutaneous endoscopic lumbar discectomy; SNA, sinuvertebral nerve ablation.

Results

The study included 136 patients (83 males, 53 females; mean age, 61.6 years; range, 51 to 75 years). LDH was found at L4/5 in 70 cases and at L5/S1 in 66 cases. Among the patients, 48 cases of LDH herniation were located on the left side of the spinal canal, 48 cases were located on the right side of the spinal canal, and 40 cases were located in the anterior part of the spinal canal. The patients were grouped based on whether SNA was performed. Patients who received SNA were assigned to the PSNA group (68 patients), and patients who did not receive SNA were assigned to the PELD group (68 patients). All 136 patients completed the surgery and achieved good postoperative recovery. All patients were followed up for at least 12 months postoperatively. Table 1 displays the baseline data, such as sex, age, LDH segment, and location of disc herniation between the two groups, showing no statistically significant differences (all p>0.05) (Table 1).

Table 1

Comparison of baseline characteristics between two groups

Following verification, the two groups of data conformed to a normal distribution and the assumption of sphericity. No significant interaction was observed between group and time. Nonetheless, significant differences in intergroup VAS scores for LBP and leg pain, JOA scores, and ODI were observed. Changes in VAS scores for LBP and leg pain, JOA scores, and ODI indices across time points were statistically significant but were not entirely consistent.

Compared with the preoperative scores, the VAS scores for LBP at 1, 3, and 12 months postoperatively were significantly lower (p<0.05). The results revealed that the VAS scores in the PSNA group were significantly lower than those in the PELD group at 1,3, and 12 months postoperatively (p<0.05) (Table 2).

Table 2

Comparison of VAS scores for LBP between two groups before and after surgery

Compared with the preoperative scores, the VAS scores for leg pain at 1, 3, and 12 months postoperatively were significantly lower (p<0.05). The results indicated that the VAS scores in the PSNA group were significantly lower than those in the PELD group at 1, 3, and 12 months postoperatively (p<0.05) (Table 3).

Table 3

Comparison of VAS scores for leg pain between two groups before and after surgery

Compared with the preoperative scores, the JOA scores at 1, 3, and 12 months postoperatively were significantly improved (p<0.05). Notably, the JOA scores were significantly higher in the PSNA group than in the PELD group at 1, 3, and 12 months postoperatively (p<0.05) (Table 4).

Table 4

Comparison of JOA scores between two groups before and after surgery

Compared with the preoperative scores, the ODI scores at 1, 3, and 12 months postoperatively were significantly decreased (p<0.05). The ODI scores were significantly lower in the PSNA group than in the PELD group at 1, 3, and 12 months postoperatively (p<0.05) (Table 5).

Table 5

Comparison of ODI between two groups before and after surgery

At the final follow-up assessment, the modified MacNab criteria showed no significant difference in clinical efficacy between the groups (p>0.05). According to the modified MacNab efficacy evaluation standard, the clinical excellent and good rates were 97.06% in the PSNA group (36 cases, excellent; 30 cases, good; and three cases, fair) and 92.65% in the PELD group (30 cases, excellent; 33 cases, good; and five cases, fair).

Discussion

MCs, as a form of pathological alteration in the endplate, are correlated with LBP [12]. Increased innervation is found in the vertebral endplates of patients with LBP and MCs, and pathological changes in this region may cause LBP. The pain caused by MCs is transmitted through the sinuvertebral nerve [13]. Patients with MCs are now considered a separate group [14,15]. Our results revealed that patients in both groups experienced a significant reduction in postoperative LBP and leg pain, as well as a significant reduction in VAS scores, compared to preoperative levels. However, some patients experience residual symptoms after PELD. Approximately 15% to 25% of patients experience recurrent LBP within 2 years after undergoing discectomy [16], and some patients may require a second surgery [17,18]. In particular, patients with pathological endplate changes may experience residual LBP after discectomy during long-term follow-up [19]. MCs have been proven to be a significant risk factor for chronic LBP after discectomy [20]. A previous study reported that patients with MC I may have persistent residual LBP after PELD, which may even worsen [21,22]. The authors suggest that this may be related to untreated pathological endplates [10].

Pain signals from damaged vertebral endplates are transmitted via the sinuvertebral nerve [23,24]. Ablation of the basivertebral nerve, a branch of the sinuvertebral nerve, has demonstrated promising clinical efficacy in patients with chronic LBP [25]. However, the widespread clinical application of SNA is limited by adverse events. Recent research has demonstrated that SNA can effectively treat LBP [8,26]. Fischgrund et al. [27] conducted an extensive long-term follow-up study on patients after SNA surgery, spanning an average duration of 6 years. The results indicate that SNA can alleviate chronic LBP in patients with MCs. Conger et al. [28] have demonstrated that SNA is a safe, effective, and minimally invasive method for the treatment of intractable LBP, consistent with the findings of our study. Although the course of sinuvertebral nerve remains controversial, anatomical studies suggest that its origin is close to inferior margin of the pedicle, providing a theoretical basis for clinical treatment [29,30]. Kim et al. [26] recommend setting the radiofrequency ablation zone at the junction of the ipsilateral articular process joint and the horizontal line of the intervertebral disc. Li et al. [8] suggest performing the surgery at the ruptured site of the posterior 1/3 of the annulus fibrosus of the intervertebral disc. Kim et al. [9] located the target site of SNA in the area between the posterior margin of the annulus fibrosus and posterior longitudinal ligament. Based on our previous research, most sinuvertebral nerves are distributed as shown in Fig. 3.

Fig. 3

The course of the sinuvertebral nerve (posterior view). The red curve shows the course and distribution of the sinuvertebral nerve and its branches, while the blue area indicates the planned surgical range for each operation. ^a)Transverse process. ^b)Pedicle. ^c)Vertebral body. ^d)Posterior longitudinal ligament. ^e)Intervertebral disc. ^f)Nerve root.

Regarding the target area for surgical radiofrequency ablation in this study, the inferior edge of the vertebral arch was the superior boundary, the level of intervertebral disc rupture was the inferior boundary, the inner edge of the fibrous ring rupture near the midline was the medial boundary, and the inner edge of the intervertebral foramen was the lateral boundary. The red curve shows the course and distribution of the sinuvertebral nerve and its branches, whereas the blue area indicates the planned surgical range for each operation, which may be adjusted according to specific circumstances. As shown in Fig. 3, SNA can sever at least one sinuvertebral nerve and its branches in the surgical segment, thereby achieving the surgical objective. Our research center does not have the qualifications to perform basivertebral nerve ablation. Nonetheless, this study provides a method to relieve pain using a minimally invasive approach in LDH patients with MC I.

The current study compared the VAS scores for back pain, VAS scores for leg pain, ODI, and JOA scores before surgery and at 1, 3, and 12 months postoperatively. Postoperatively, all clinical outcomes improved. These scoring indicators showed that PELD is an effective surgical method for the treatment of LDH, significantly alleviating pain, improving functional status, and enhancing long-term outcomes. Patients who received SNA exhibited superior clinical outcomes in both short-term and long-term follow-ups. Additionally, a study suggested that epidural neovascularization with adhesion reflects aberrant neurological connections [28], which are associated with reflex inhibitory mechanisms of the multifidus muscle, thereby inducing spasm and leading to spasmodic lumbar pain. During the surgical procedure, SNA can be combined with radiofrequency ablation of connective tissues, such as epidural neovascular adhesions, to alleviate LBP caused by paravertebral muscle spasms, ultimately contributing to the relief of postoperative back pain.

We observed a significant reduction in leg pain in the PSNA group, which we believe may be attributable to the following reasons. The surgical goal of PELD is to remove the nucleus pulposus and reduce nerve root compression. Owing to the requirements of SNA surgery, the actual surgical scope in the PSNA group was larger than that in the PELD group according to preoperative planning. During the operation, additional manipulation of the tissues surrounding the dural sac and nerve roots was also performed. In terms of the actual decompression achieved, the PSNA group demonstrated a greater reduction than the PELD group. Consequently, as mechanical compression of the nerve roots was more notably alleviated, patients in the PSNA group experienced more pronounced relief from leg pain. Abnormal proliferation of various soft tissues surrounding the nerve root may cause local adhesions and abnormal stimulation of the nerve root. During SNA surgery, these proliferated abnormal tissues can be scorched, allowing extensive release of the nerve root, which may also alleviate pathological pain.

Nevertheless, the limitations of this study should be acknowledged. The lack of disc herniation recurrence may be attributed to the relatively small sample size and short follow-up period. Moreover, the single-center design, potential observer bias, and lack of randomization are further limitations. Additionally, this study included only patients with single-segment LDH and MC I. Currently, basivertebral nerve ablation is not performed at our research center; therefore, it was not discussed. Future research should aim to expand the sample size, adopting a multi-center, prospective, randomized controlled study design to investigate the impact of PELD and SNA in the treatment of LDH with MC I.

Conclusions

In summary, our findings demonstrate that PELD is a safe and effective surgical approach for the treatment of LDH with MC I and that the combination of SNA during surgery can enhance surgical efficacy, leading to better long-term outcomes.

Data Availability

The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.

Article information Continued