Clinical Outcomes of Full-Endoscopic Disc Cleaning Surgery for Chronic Low Back Pain Associated With Modic Changes: A Case Series
Article information
Abstract
This article reports a case series evaluating the short-term clinical outcomes of full-endoscopic disc cleaning (FEDC), which is one of the most minimally invasive surgical procedures for treating chronic low back pain (CLBP) due to Modic changes (MCs). Seven patients who underwent FEDC at Tokushima University Hospital between 2019 and 2021, with postoperative clinical outcomes available over a follow-up period of more than 12 months, were included in the study. Six men and one woman underwent FEDC. The average age of the patients was 48.7 years with an average CLBP duration of 9.4 years. Five of these patients showed favorable outcomes, with reductions in LBP. The other 2 required revision surgery at the 12-month follow-up due to recurrent symptoms. One of the patients who required revision surgery was the oldest in this series, a 73-year-old woman with mixed signal changes including Modic type 3 at the anterior corner at L2–3 to L4–5 levels with spinal malalignment. Although disc block was immediately effective, pain reproduction on discography yielded unclear results. The clinical outcomes of this series suggest that FEDC may be particularly effective for patients with inflammatory type 1 and type 2 changes and less advanced degenerative changes. However, in cases involving more complex degeneration, such as type 3 changes or notable spinal deformities, CLBP may arise from multiple contributing factors beyond MCs alone, underscoring the importance of a thorough evaluation when determining surgical indications for FEDC.
INTRODUCTION
Modic changes (MCs), first described by Michael Modic in 1988, represent pathological alterations in the vertebral endplates and subchondral bone marrow, typically associated with intervertebral disc degeneration. These changes are detected through magnetic resonance imaging (MRI) and classified into 3 types: type 1 (inflammatory changes), type 2 (fatty replacement of bone marrow), and type 3 (sclerotic changes in the vertebrae) according to the change of T1 weighted and T2-weighted images [1,2]. Among these, type 1 MC have been most strongly correlated with chronic low back pain (CLBP) and are thought to represent an ongoing inflammatory process [3,4].
MCs are prevalent in individuals with degenerative spinal disorders, affecting up to 40%–50% of patients with LBP [5]. The incidence of MCs increases with age and disc degeneration, with type 2 MCs predominating [6,7].
Conservative (nonoperative) treatments for CLBP associated with MCs have included a variety of approaches such as physical therapy, medications such as nonsteroidal anti-inflammatory drugs (NSAIDs), opioid, and bisphosphonate, and steroid injection [8]. In some cases, antibiotic therapy has been proposed, based on the hypothesis of low-grade bacterial infection, potentially driven by Propionibacterium acnes in the disc space, although this remains controversial due to inconsistent clinical results [8-10]. However, the outcomes are often suboptimal, particularly in patients with type 1 MC due to the inflammatory nature of the pathology, many patients continue to suffer from refractory pain, prompting the need for more definitive interventions [8].
For patients whose symptoms persist despite exhaustive conservative measures, surgical options such as spinal fusion have been traditionally utilized. Although anterior interbody fusion had slightly better results than posterolateral fusion, both groups had better results with or without intradiscal procedure, suggesting that LBP related to MCs is influenced by excessive stress and instability in the affected segment [11]. However, it comes with risks such as adjacent segment disease, prolonged recovery, implant failures, and surgical site infection.
The development of minimally invasive spine surgery, especially full-endoscopic spine surgery, has been remarkable in recent years, including intradiscal treatment for discogenic LBP [12,13]. Full-endoscopic disc cleaning (FEDC), in which the degenerative discs are removed, the endplate with MCs is radiofrequency ablated, and the disc space is cleaned at the same time as degenerative disc excision, has been reported [13,14]. This technique aims to reduce the inflammatory mediators and ablate pathological neovascularization and grown nerve of the vertebral endplate, thereby alleviating pain. Although this technique offers several advantages, including reduced operative time, faster recovery, and decreased postoperative complications for MCs-related LBP, there are only a few reports on the postoperative clinical outcomes after FEDC. In this study, we show the case series of short-term outcomes in FEDC for the patients with CLBP due to MCs.
MATERIAL AND METHODS
1. Inclusion Criteria
This study included adult patients who were diagnosed with CLBP due to MCs and underwent FEDC at our hospital between September 2019 and January 2021. Patients with a follow-up period of 12 months or more were included in this study. Patients exhibiting radicular pain in the lower extremities or buttocks associated with spinal canal or foraminal stenosis were excluded in this study. All surgeries were performed by the same single surgeon (KS). For cases that required revision surgery after FEDC, the follow-up period was defined as the duration until the revision surgery was performed. LBP was assessed using the visual analogue scale (VAS), which rates pain on a scale of 0 to 10. VAS of LBP was measured preoperatively, and at postoperative 1 month, 3, 6, 12, and 24 months. This study was approved by the ethics committee of Tokushima University, and written informed consent was not required, as approved by the ethics committee (approval No. 3642).
2. Surgical Indications
The diagnosis of LBP due to MCs was confirmed by MRI findings, followed by discography and block at the level with MCs. Discography was performed under fluoroscopic guidance until the contrast medium spread throughout the entire intervertebral space, and we checked the presence of reproductive pain during discography. Subsequently, approximately 2 mL of 1% lidocaine was injected into intervertebral space for the block. Patients who experienced relief of LBP after the block were diagnosed with MCs-related LBP. After diagnosis, conservative treatment, including administrations of NSAIDs and antibiotics, and physical therapy was continued for more than 3 months. Patients whose LBP recurred and persisted were considered as candidates for FEDC.
3. Surgical Procedure
As the preoperative procedures, computed tomography (CT) scans after discography and block were conducted in the prone position to confirm the site of skin incision and the trajectory angle in the operation.
All surgical procedures were performed in prone position under local anesthesia with 15–20 mL of 1% lidocaine (depending on the patient's body size) using walking technique through a transforaminal approach [14]. In brief, an 8-mm skin incision was made, and the full-endoscope was inserted to the lumbar foramen after local anesthesia and discography. Using the foraminoplastic outside-in technique, sufficient foraminoplasty was performed to prevent irritation or injury to the exiting nerve root (ENR). After the sufficient disc was removed from the outside to decrease the intradiscal pressure, the endoscope moved inside the disc. Degenerated disc tissue was excised, and the vertebral endplate was ablated as thoroughly as possible. In cases of MCs-related LBP, degenerative discs often appeared red or brownish in color, and pathological hemorrhagic microvasculalization was observed at the vertebral endplate (Figure 1). The disc space was then thoroughly irrigated with saline. These procedures inhibit the influx of inflammatory cytokines to degenerative endplate and, at the same time, causes the denervation of the pathologically grown nerves that accompany the neovascular into the vertebral endplate [15]. A closed-suction drain was placed at the foramen, and the surgery was completed.
Representative endoscopic findings during full-endoscopic disc cleaning surgery. (A) The endoscopic view after the endoscope was inserted into the disc. Yellowish-brown annulus fibrosus and denuded endplate cartilage (black arrows) can be seen around the nucleus pulposus stained blue by indigo carmine. (B) The endoscopic view after the resection of degenerative discs. Because pathological neovascularization into the vertebral endplates causes hemorrhage, the lesions of hemorrhage and reddish degeneration (yellow arrows) are ablated by radiofrequency as extensively as possible.
4. Postoperative Treatment
Postoperatively, patients were kept on bed rest for 2 hours. After we confirmed their general condition was stable, they were fitted with a thoracolumbar orthosis, primarily a Jewett brace, and allowed to start ambulation. The drain was removed the following day. NSAIDs and antibiotics were administered by oral intake for 4 weeks postoperatively, and patients were advised to wear the thoracolumbar orthosis and restrict activities to a minimal daily activities and light desk work. After this period, dosage of NSAIDs was tapered according to pain levels, and patients were gradually encouraged to increase their activity levels without restrictions.
5. Evaluation of Images
1) Classification of MCs
The subendplate changes at the affected level were classified as type 1 to type 3 based on the signal changes in T1-weighted and T2-weighted images, as previously described [1.2]. Additionally, short tau inversion recovery (STIR) images were obtained to assess the presence of signal changes at the lesions of MCs to show active inflammation or bone marrow edema clearly [16].
2) Lumbar alignment
Preoperative standing radiographs of the lumbar spine were used to evaluate the lumbar lordosis (LL) angle between cranial endplate of L1 and S1 vertebrae, and local Cobb angle between cranial endplate of cranial vertebra and caudal endplate of caudal vertebra at the affected level with MCs.
RESULTS
Seven cases were included in this study (Table 1). There were 6 male and 1 female patients, with a mean age of 48.7±6.58 years (range, 32–73 years). The mean body mass index was 23.6±0.6 kg/m2 (range, 20.9–25.7 kg/m2), and the mean duration of CLBP before diagnosis at our hospital was 9.4±1.3 years (range, 4–15 years). The most commonly affected level was L5–S1, involving 5 levels, followed by L4–5 in three levels and L3–4 in one level, including 2 cases of multilevel involvement. The most major MCs near the endplate were classified as type 1 in three levels, type 2 in three levels, and type 3 in 1 level. High signal change on STIR images at the same lesions to the MCs near the endplate was positive in 6 cases and negative in 1 case. The LL angle was 38.5°±5.3° (range, 11.6°–52.6°), and the local Cobb angle was 2.0°±1.2° (range, 0.1°–8.7°). Reproductive pain during discography was positive in 5 cases and negative in 2 cases.
Characteristics of patients who underwent transforaminal full-endoscopic disc cleaning under local anesthesia
No intraoperative or postoperative complications, such as headaches, nausea, nerve root injuries, dural tears requiring repair, and surgical site infections, or epidural hematomas, were observed during the perioperative period.
The preoperative VAS of LBP was 6.3±0.5 (range, 5–8). VAS of LBP improved to 2.9±0.7, 3.1±1.3, 3.4±1.3, and 3.1±1.4 at 1 month, 3 months, 6 months, and at the final follow-up, respectively, showing improvement that was sustained after 3 months postoperatively (Figure 2).
Postoperative clinical course of the visual analogue scale (VAS) for low back pain (LBP) after full-endoscopic disc cleaning in each case. All 7 cases showed improvement in the VAS for LBP at 1 month postoperatively, but at 3 months postoperatively, cases 4 and 5 experienced recurrence of LBP, with a VAS score similar to that observed preoperatively. These patients required revision surgery at 12 months postoperatively. Data are presented as mean ± standard error of the mean.
Two of the 7 patients required revision surgery due to persistent LBP at 12 months postoperatively (one case underwent rhizotomy, and the other underwent a revision FEDC at the same level).
1. Presentation of Failure Case (Case 4)
A 73-year-old woman with a 7-year history of CLBP. Her initial VAS of LBP was 7 out of 10, with severe pain upon rising in the morning and during forward bending (e.g., while washing her face), extension, and short periods of walking. No neurological abnormal findings were observed in the buttocks or lower extremities. Standing plain radiographs of the lumbar spine revealed degenerative malalignment with mild scoliosis with right lateral wedging at L3–4 level and decreased LL angle (Figure 3). MRI revealed MCs at L2–3 to L4–5 levels, particularly apparent signal changes at L3–4, where the anterior corner of the endplate showed type 3 MC, and the surrounding area of type 3 exhibited type 1 MC at L3–4 and type 2 MC at L2–3 and L4–5 levels (Figure 4). STIR images showed the highest signal change at L3–4 level around low signal at the anterior corner of endplate (Figure 4). Because of tenderness in the back at the L3–4 level, discography and a block were performed at this level (Figure 5). Although reproductive pain during contrast medium injection was unclear, the block provided sudden pain relief, leading to the diagnosis of LBP related MC at L3–4. After the block, her pain was temporarily relieved with the use of a Jewett brace and NSAIDs, but it recurred.
Preoperative standing plain radiographs of the lumbar spine. (A) Anteroposterior (AP) view showing a mild degenerative scoliosis with endplate bony spar formations at multiple levels and the right lateral wedging at L3–4 level. Yellow dot lines show the local Cobb angle. (B) Lateral view in neutral position showing decreased lumbar lordosis with sclerotic changes in the anterior part of endplate and vacuum phenomenon at L2–3 to L4–5 levels.
Preoperative para-median sagittal magnetic resonance imaging (MRI) of lumbar spine. (A) A left T1-weighted image showing low signal changes in the anterior corner of endplates at L2–3 to L4–5 levels and high signal changes around the low signal change at L2–3 and L4–5 levels (yellow arrows). (B) A middle T2-weighted image showing low signal changes in the anterior corner of endplates at L2–3 to L4–5 levels and high to iso signal changes around the low signal changes at the same levels (yellow arrows). (C) A right short tan inversion recovery (STIR) image showing low to iso signal changes in the anterior corner of endplates at L2–3 to L4–5 levels and high signal changes around the low signal change at L2–3 and L3–4 levels (yellow arrows). The L3–4 lesion is the most apparent.
Preoperative discography and block at L3–4 level. (A) An anteroposterior (AP) and lateral views (B) showing the radiographs during discography. The entire intervertebral space was contrasted by injecting 2 mL of contrast medium. There was not apparent productive pain during discography.
FEDC was performed as described in material methods section. VAS of LBP improved to 2 from 7 at 1 month postoperatively; however, pain gradually recurred starting at 3 months postoperatively, with VAS of LBP persisting at 8–9 up to 12 months postoperatively. Considering the possibility of facet joint-related pain at multiple levels, she thereafter underwent revision surgery for rhizotomy.
DISCUSSION
1. Diagnosis of Low Back Pain due to MCs
MRI findings are crucial in diagnosing LBP due to MCs. Many reports have highlighted the relationship between type 1 and type 2 MCs and LBP [1,2]. However, there are some reports on the weak relationship between type 3 MCs, LBP, and treatment outcomes [7,17]. Since type 3 is considered an advanced stage of degeneration following types 1 and 2, it is possible that the cause of LBP in these cases is not limited to the vertebral endplates but may involve more complex factors. In this case series, case 4 had type 3 changes in the anterior corner of the vertebral endplate at L2–3 to L4–5 levels. This patient was the oldest case and exhibited the most severe lumbar malalignment in this case series, suggesting that her LBP may have been caused by multiple factors. Advanced degenerative changes and malalignment with type 3 change of the vertebral endplate may not be ideal candidates for FEDC.
Additionally, in this case series, pain relief from disc blocks was used as a diagnostic method for LBP due to MCs. Both pain reproduction during discography and pain relief from blocks have been traditionally reported as useful diagnostic methods for discogenic low back pain [18,19]. However, false positives may occur with discography, as even normal discs can produce pain, and patients may struggle to differentiate between the nonspecific pain by discography and their reproductive pain like usual pain patients complain of. Furthermore, disc blocks can spill over the injected anesthetic around the nerve roots or epidural space, potentially causing radicular pain to be misdiagnosed as disc- or endplate-related LBP. In this case series, pain reproduction during discography was unclear in cases 4 and 6, with case 4 requiring revision surgery and case 6 showing improvement in the VAS of LBP from 8 to 4 (50% improvement rate) but still having the highest pain score at final follow-up excluding the 2 revision cases. While disc blocks were used to diagnose LBP at the affected level with MCs in this case series, a certain rate of false positives must be considered. Therefore, cases with both apparent pain reproduction during discography and pain relief after disc block may be more appropriate candidates for FEDC.
2. Minimally Invasive Surgery for LBP due to MCs
Traditionally, lumbar fusion surgery has been widely performed for LBP originating from the disc and vertebral endplate lesions, with proven efficacy [11]. However, fusion surgery carries risks such as prolonged restriction of activities, implant loosening or breakage, pseudarthrosis, and adjacent segment disease.
Recently, minimally invasive surgeries for disc- and endplate-related LBP, including MCs, have been developed [13,14].
A similar report has been published by Kim et al. [20,21], describing a method for ablating the basivertebral nerve and sinuvertebral nerve using an epiduroscopic technique under full-endoscope. In their report, decompression of bulging intervertebral discs and stenotic regions of the lumbar spinal canal is performed at the beginning of the procedure. Therefore, the indications in their study differ slightly from those in our report, which focuses on cases without stenotic findings. Nevertheless, their study demonstrated favorable short-term outcomes for CLBP associated with MCs and disc degeneration through peripedicular ablation.
In addition, reports on basivertebral nerve ablation (BVNA) through transpedicular approach have emerged [22]. The indications and complications for BVNA and FEDC are similar, there is a difference of trajectory between FEDC and BVNA. FEDC allows for an approach through the disc, making it possible to address pain from degenerative discs with high-intensity zones (HIZ) at the same time [14]. Intraosseous BVNA involves manipulations within the vertebral body or pedicle, requiring caution for vertebral or pedicle fractures, whereas FEDC involves the removal of degenerative disc material while approaching the vertebral endplate via a transforaminal approach. Therefore, there is a risk of disc height reduction or progression of degeneration after FEDC, making long-term follow-up with MRI advisable.
3. Challenges of the FEDC Procedure
One of the challenges associated with the FEDC procedure, as mentioned above, is the potential impact on the intervertebral disc postoperatively. The surgical technique involves ablation and removal of the degenerated disc along the pathway, followed by ablation of the vertebral endplate surrounding the MCs. Consequently, in cases without concurrent disc pathologies such as herniation and HIZ in the posterior annulus fibrosus, the removal and ablation of the nucleus pulposus may result in excessive invasion of the disc. This could potentially promote postoperative disc degeneration and a reduction in disc height.
However, diagnostic tools useful for identifying CLBP associated with MCs, such as discography and disc block, are typically performed via a transdiscal approach. It is often challenging to strictly differentiate between LBP originating from vertebral endplates, such as MCs, and degenerated disc associated with such as HIZ. When MRI reveals both endplate and disc abnormal findings in the same segment, a combined pathophysiology involving discogenic pain should also be considered, and such cases are common. For cases where symptoms are strictly attributed to the vertebral endplates, a transpedicular or peripedicular procedures targeting the basivertebral nerve may offer a less invasive treatment option for the disc. However, FEDC allows simultaneous treatment of both the disc and vertebral endplates via a transforaminal approach, enabling sufficient intervention for both lesions, which is considered an advantage [14].
Regarding concerns about postoperative disc height reduction, partial removal of the nucleus pulposus makes it a challenging issue to completely avoid in FEDC. Nevertheless, patients with CLBP accompanied by MCs often already exhibit advanced disc degeneration at the affected levels, with a decreased disc height. Therefore, the likelihood of excessive removal of healthy nucleus pulposus leading to significant height reduction may not be high. However, for cases with relatively preserved nucleus pulposus, careful consideration of indications for FEDC procedures is warranted.
Meanwhile, FEDC procedure involves sufficient foraminoplasty before accessing the disc. As a result, the foraminal area is expanded compared to the preoperative state. This expansion may help mitigate the risk of ENR irritation or injury caused by intraoperative manipulation or mild postoperative disc height reduction.
CONCLUSION
FEDC using a transforaminal approach under local anesthesia is a minimally invasive technique with low risk of complications and enables us direct procedures to the affected intervertebral discs and vertebral endplates in patients with intractable LBP due to MCs. This technique has allowed patients to avoid spinal fusion and return to daily activities in a short period of time. However, we experienced 2 cases in which LBP recurred postoperatively and required revision surgeries. Future studies will include a larger sample size and more detailed analysis of factors affecting surgical outcomes.
Notes
Conflicts of interest
The authors have nothing to disclose.
Funding/Support
This study received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.