Percutaneous transforaminal endoscopic decompression with removal of the posterosuperior region underneath the slipping vertebral body for lumbar spinal stenosis with degenerative lumbar spondylolisthesis: a retrospective study

In patients with LSS with DLS, most experts currently believe that conservative treatment failure for at least 3–6 months is an indication for surgery [18]. Two surgical methods are commonly used nowadays, namely decompression alone and decompression combined with fusion. It is widely recognized that patients with unstable DLS, especially those with obvious axial pain, will benefit from fusion [19]. Several studies have shown that there are two criteria for selecting decompression alone in a population of patients with DLS and LSS: (1) mainly characterized by symptoms of radicular pain and neurogenic claudication [20]; (2) stable DLS (sagittal translation ≤ 3 mm on standing extension-flexion radiographs radiograph or angular displacement ≤ 10°) [12, 13]. Therefore, for patients with stable DLS and LSS with neurological symptom, the main purpose of surgery is to relieve nerve compression, and vertebral fusion and fixation are not necessary. Decompression without fusion has many advantages, including reduced complexity of surgery, short hospitalization time, and reduced cost [21]. Moreover, because patients with LSS and DLS are usually old and weak and have comorbidities, the open instrument fusion decompression method performed under general anesthesia is associated with high perioperative risks. Therefore, it is important to consider the necessity of instrument fusion in elderly patients with comorbidities. Recently, with the development of spinal endoscopy, PTED has been considered an alternative to fusion surgery due to its cost-effectiveness, low complications, reduced surgical time and blood loss [22], and minimal impact on spinal stability [14]. A summary of current state of research on transforaminal approaches for DLS is presented in Table 4.

DLS and LSS can elicit three different types of pain through different mechanisms [23]. The first type is mechanical lower back pain, and the main cause of this pathology is the degeneration of the intervertebral discs and facet joints or lumbar spine instability. The second type is neurogenic claudication, which is caused by concomitant CSS secondary to DLS as well as hypertrophy of the LF and osteophytes due to facet arthrosis invading the spinal canal. The third type is pain due to compression of the nerve roots in the lateral recess or foramen caused by DLS. We explored the second and third types of pain in this study, and the main symptoms of the patients were neurogenic claudication or root pain caused by LSS combined with low-level stable DLS. Therefore, the main goal of surgery was to alleviate nerve root symptoms and neurogenic claudication and to decompress LSS exacerbated by DLS. We introduced the procedure of removing the PRSVB during PTED, which can expand the lumbar spinal canal without excessive small joint resection. We observed significant improvements in the VAS and ODI scores at 3 months, 6 months, and 12 months postoperatively. According to the modified MacNab criteria, the “excellent” and “good” rating was reported in approximately 93.2% of patients, which was better than that reported with other endoscopic treatments [24, 25]. Furthermore, the incidence of surgery-related complications (6.8%) in this study was not significantly higher than that in previous endoscopic studies [16, 26]. And 12 months after surgery, all patients underwent MRI to confirm no recurrence. Thus, it was preliminarily indicated that PTED combined with removing the PRSVB elicited a good clinical effect in patients with DLS and LSS.

The PRSVB caused by DLS can be combined into three pathological features based on the basic pathological conditions of LSS. The first is the most common case related to lateral recess stenosis [16], which is usually caused by disc herniation, hypertrophy of the LF and articular processes. The uneven PRSVB may combine with these factors to exacerbate the compression of nerve root. The second case is related to CSS. Hypertrophic articular processes and LF and hyperplastic osteophytes often invade the spinal canal in patients with severe LSS The uneven PRSVB that protrudes into the spinal canal further reduces the space of the spinal canal, leading to the occurrence of CSS and aggravating claudication [14]. The reduction in effective activity space of the cauda equina nerve and traversing nerve roots exacerbates claudication. The third case is related to intervertebral foramen stenosis. In patients with advanced DLS, PRSVB is the main factor causing intervertebral foramen stenosis due to anatomical variation and the decrease of intervertebral foramen height [26]. Thus, any factor that causes LSS may combine with PRSVB formed by DLS, and the aforementioned pathological conditions can exist alone or in combination to worsen the symptoms of the patient.

Jasper et al. [25] reported that among the 21 patients with DLS who had undergone PTED, 3 (11.9%) relapsed in the first 3 months. The relapsed patients actively underwent enlargement of the foramen and resection of the PRSVB during revision surgery, and they recovered well postoperatively. This indicated that to achieve comprehensive decompression, attention must be paid to the removal of the PRSVB. To make decompression efficient and thorough on the dorsal and ventral sides, we used bone drills in the first foraminoplasty, which enlarge the foramen and remove the LF efficiently and safely [24]. Notably, the presence of PRSVB reduces the effective space in the spinal canal, causing the nerve roots and dura mater to be squeezed and deformed. The position of the bone drill should not blindly pursue the midline of the spine, as it could easily damage the dura mater and nerve roots. When the patient feels pain, the orthopedic surgeon should withdraw the bone drill in time and insert the working channel directly to perform foraminoplasty endoscopically. We introduced a microscopic bone knife during the second foraminoplasty, which can remove the bony structure blocking the field of vision as needed, and more importantly, it can flexibly remove the PRSVB. After 270° decompression of the ipsilateral nerve root, the abduction angle of the endoscope needs to be increased to decompress the contralateral central spinal canal and nerve root. The skillful removal of the PRSVB lies in carefully removing it bit-by-bit using the microscopic bone knife and taking it out using small straight forceps in time. The pathological changes of LSS with DLS are slow and complex until obvious symptoms appear. The nerve roots and dura mater remain compressed for a long time, and local inflammation may lead to scar adhesion. Therefore, it is necessary to distinguish the anatomical structure of the intervertebral space carefully during surgery before performing decompression operations. The adhesion between the dural sac, nerve root, and protrusion should be first loosened to reduce the risk of nerve and dural injury. Additionally, because osteoporosis patients are prone to bleeding and a blocked surgical field of vision, it is necessary to control bleeding strictly through radiofrequency coagulation and sufficient irrigation pressure. In patients with neurogenic claudication with unilateral or bilateral leg symptoms, the PRSVB compressing the traversing nerve roots and cauda equina must be removed to expand the central spinal canal, lateral recess, and foramen. Postoperative CT can show that the ventral and dorsal edges of the spinal canal are aligned and becoming smooth and continuous, which is similar to the postoperative findings of the reduction method in open surgery. We have created a model diagram to illustrate the key points further (Fig. 7).

In this study, the patients exhibited no significant progression of slippage during the last follow-up compared to that present preoperatively, indicating that PTED will not affect the natural process of LSS with DLS, which corroborates previous reports [14, 16, 20]. Ahuja et al. [27] found that the removal of more than 30% of the facet joint would increase spinal mobility and the pressure in the intervertebral disc. Foraminoplasty aims to remove the bone effectively from the ventral side of the SAP and protect the integrity of the facet joint. In this study, our technique only removed a small part of the facet joint, accounting for less than 10–20% of the total facet joint. Moreover, removing the PRSVB can avoid excessive removal of the back of the facet joint. The imaging results of this study showed that sufficient decompression and facet joint preservation were achieved in almost all patients. During the postoperative follow-up, no patients showed any further clinical or radiological segmental instability on dynamic radiographs; hence, this decompression technique may not produce obvious iatrogenic spinal instability.

Currently, there are many endoscopic decompression methods for LSS with DLS, such as percutaneous endoscopic interlaminar decompression [28], unilateral biportal endoscopic decompression [29], and microendoscopic decompression [30]. All these techniques adopt a posterior approach, providing surgeons with a friendly and familiar view and less tissue damage. The main concept of these methods is dorsal decompression, requiring general anesthesia and partial posterior joint resection. However, it is possible that the ventral PRSVB cannot be completely removed, the decompression of the contralateral nerve root is difficult, the decompression of foraminal stenosis is limited, and the surgical risk increases in elderly patients under general anesthesia. Additionally, elderly patients with LSS and DLS often have stenosis of the interlaminar space due to hyperplastic osteophytes and ossified LF; hence, posterior decompression to remove the dorsal bone and LF may be time-consuming. Excessive dorsal resection and decompression may lead to fractures and segmental instability [31], and the risk of dural injury increases.

The PTED technique used in this study is superior to the posterior approach in the degree of ventral decompression achieved. The PRSVB formed by DLS is an important pathological factor for CSS, lateral recess stenosis, and foraminal stenosis. PTED not only achieves 270° decompression of the nerve root but also removes the PRSVB with a good visual angle. When this part is removed, the area of the spinal canal is enlarged to relieve claudication [5]. Moreover, the enlargement of the ventral space ensures that the endoscope has enough space to pass in front of the dura mater to complete the decompression of the contralateral nerve root. Ultimately, the ideal surgical endpoint is identified by observing the entire circular fissure, epidural pulsation of the dural sac, and free movement of the nerve roots. Therefore, PTED combined with the removal of PRSVB has advantages in achieving decompression in patients with LSS and DLS. Notably, surgeons should evaluate the activity of the nerve root and decompress the ventral part of the dural sac repeatedly, including the epidural space and intervertebral disc. Although some patients with LSS and DLS experience unilateral lower limb symptoms, contralateral prophylactic decompression is still necessary when performing unilateral PTED. For patients with obvious symptoms in both lower limbs, the approach side should be determined based on imaging characteristics and patient signs, and bilateral PTED may have to be considered in some cases.

The average age of patients in this study was 69.5 years, and 84.1% (37/44) of them had comorbidities. The average age and incidence of preoperative comorbidities were higher in this study than in other endoscopic reports [26]; the VAS and ODI scores of the patients in this study improved significantly during follow-up. These excellent results are not only attributed to the removal of the PRSVB technique we introduced but also to the advantage of using local anesthesia in PTED. Additionally, the average operation time in this study was 67.8 min, which was shorter than that of open fusion surgery [32, 33]. Therefore, the use of local anesthesia provides surgical opportunities for elderly patients who refuse open fusion surgery or are weak, reduces the occurrence of complications related to fusion surgery and general anesthesia, and requires simple postoperative management. Local anesthesia can provide surgeons with nervous feedback; hence, the doctors can promptly cease activities that stimulate the nerve root to prevent further injury. From the point of view of population health, PTED is a treatment method with low medical costs and a relatively low risk of complications, which is beneficial to patients with DLS and LSS, who are the first choice for PTED treatment.

Endoscopic lumbar fusion surgery, as a reliable minimally invasive spinal technique, has been used to treat lumbar degenerative diseases [34]. Some scholars have reported that full-endoscopic trans-Kambin’s triangle lumbar interbody fusion was applied to the treatment of DLS with LSS, and achieved good results [35, 36]. This technique can not only remove the disc material under endoscope, but also achieve fixation or correction of spondylolisthesis through the implantation of pedicle screws and fusion cages. Therefore, for patients with potentially unstable or unstable DLS, fusion surgery should be given priority. However, some shortcomings of endoscopic fusion have also been reported, such as the lack of direct decompression of neural structure, residual postoperative pain and postoperative cage subsidence or migration [34]. Therefore, the choice of technology depends more on the pathological characteristics of the disease, patients’ symptoms and willingness, and cost-effectiveness. For stable DLS patients who only show symptoms related to nerve root compression, it is more reasonable and beneficial to choose PTED to relieve symptoms before fusion surgery is really needed.

In this study, there was one case of dural sac tear, which was caused by the bone knife accidentally slipping and tearing the dural sac because the surgical field of vision was seriously affected by profuse bleeding while removing the PRSVB in this osteoporosis patient. Therefore, the prerequisite for the next procedure is careful hemostasis to ensure clear vision. One patient had residue of herniation. During the revision operation, fragmented-free nucleus pulposus tissue was found at the nerve root canal, which may have been due to the lack of experience of orthopedic doctors in the early stages of the surgery, resulting in insufficient clearance of the intervertebral disc and migration of the nucleus pulposus. Therefore, repeated decompression of the epidural and intervertebral discs is necessary to prevent residual nucleus pulposus. One patient had postoperative dysesthesia, which could be attributed to the nerve root injury caused by repeated punctures and foraminoplasty [37].

This study has some limitations: Firstly, due to its strict inclusion criteria of this retrospective study, the sample size was relatively small, and the evaluation study of larger sample size should be carried out in the future. Secondly, although the short-term follow-up indicated that PTED with the removal of the PRSVB achieves good results, the long-term clinical effects and complications were unknown, and long-term follow-up was still needed. Thirdly, without a control group, the advantages of PTED combined with the resection of the PRSVB and single dorsal decompression cannot be compared and studied in detail. Fourthly, data such as the changes in the spinal canal area were not specifically assessed, as we attributed the relief of lower limb symptoms to the expansion of the vertebral canal. The relevant data will be considered in future research to illustrate the advantages of PTED with the removal of the PRSVB clearly.