The treatment of lumbar spinal stenosis continues to increase, especially in older people. The aim of lumbar spinal canal stenosis surgery is relief of sciatic leg pain and improved walking ability by decompression of the spinal canal. The classic operation involves the sequential dissection of the paraspinal muscles and spinous process, followed by removal of the interspinous ligaments [123].

Scarring of the epidural space was problematic [456789]. Scars may become clinically symptomatic [567], which can make subsequent surgery more difficult due to the postoperative adhesion between the epidural space and paravertebral muscle [10111213]. The resection of structures preserving stability may promote operation-induced segmental instability [1314]. Destruction and denervation of the dorsal branch of the spinal nerves may be detrimental to the stabilization and coordination systems [81516]. Resection of joint and soft-tissue structures in the lateral and ventral area is also required for decompression of degenerative lumbar spinal stenosis, and this occurs often.

Minimally invasive surgery that better spares tissue are being increasingly used [1017]. New surgical techniques of disc herniations in the lumbar spine now permit a full-endoscopic procedure under continuous irrigation, and is minimally invasive [1819]. One key issue is the possibility of adequate bone and ligamentum flavum resection under continuous visual control [181920]. This also enables the technique to be used in spinal stenosis surgery [20].

We use two portal percutaneous endoscopic decompression for lumbar spinal canal stenosis. This procedure has a number of advantages. It permits the use of ordinary arthroscopic and spine instruments without need for special endoscopic sets. It allows free movement and angulation of the surgical tool and the endoscope independent of each other as they are not restricted by the confines of a common working portal which results in marked reduction in the technical difficulties. Finally, the use of saline irrigation abolishes the problem of repeated cleaning of the endoscopic lens of accumulated fog or blood. This technique is useful for the minimally invasive surgical treatment of spinal canal stenosis. However, studies of this technique have been limited.

The objective of the current study was to report the outcomes of patients with lumbar spinal stenosis treated percutaneous endoscopic decompression, focusing on the surgical techniques and the results of the patients' clinical evaluations.

This study was approved by our Institutional Review Board and informed consent was obtained from each patient. Thirty consecutive patients undergoing percutaneous endoscopic decompression for lumbar spinal stenosis between January 2012 and June 2013 were enrolled. The inclusion criteria were neurogenic claudication or radicular leg pain with associated neurologic signs referring to the lumbar spinal stenosis syndrome; moderate-to-severe spinal canal stenosis shown on cross-sectional imaging, such as magnetic resonance imaging (MRI) or computed tomography (CT) scan, and failure of conservative treatment for at least 3 months. The exclusion criteria were spondylolisthesis more than Meyerding Grade I; prior surgery in the same segment; coronal plane deformity with Cobb angle more than 10 degrees and spinal instability in dynamic radiograph. The patients were positioned prone in a kneeling position after general anesthesia. The operated level was identified with image intensifier. A spinal needle was inserted and localization was rechecked with a lateral view. Endoscope portal and working portal were inserted through the two separate skin incisions and docked onto the lamina (Fig. 1). The localization was reconfirmed with lateral view of fluoroscopy before the decompression. Potential space was created with endoscopic cautery (Fig. 2). Unilateral laminectomy was performed for patients with unilateral neurological symptoms. For patients with central spinal canal stenosis symptoms, we performed unilateral approach for bilateral decompression as described by Guiot et al. [21] to decompress bilateral lateral recesses and the central canal. The ligamentum flavum was completely excised. Decompression of the lateral recess was achieved by partial facetectomy. To preserve the integrity of the facet joint as much as possible, we used high-speed pneumatic burrs with diamond heads and curved Kerrison rongeur to undercut the facet joint. Then we moved the endoscope to the central canal and contralateral lateral recess. This process was performed by gently moving over the dural sac, excising the ligamentum flavum with straight and curved Kerrison punches and undercutting the lamina with a high-speed pneumatic burr. The adequacy of decompression was determined by probing the traversing nerve roots to confirm the extent of decompression. The endpoint of decompression was the outer edges of the bilateral nerve roots. Illustrative cases are depicted in Figs. 3, 4, 5, 6. After hemostasis, no drain was placed and the incision was closed. Ambulation was allowed immediately after the surgery with a brace. Rehabilitation was started from the day after the operation. All patients were discharged from the hospital within several days of the surgery.

Preoperative and follow-up patient functions were evaluated using the Oswestry disability index (ODI), visual analogue scale (VAS) score of pain and Macnab criteria. Neurologic claudication was recorded as walking ability (<400 m, 400–1600 m, >1600 m).The results obtained were recorded in patient charts and were not the basis for any decision making. The data about the preoperative comorbidities, intraoperative and postoperative complications were retrieved from the medical chart review.

Advances in the management of lumbar spinal stenosis are being driven by the increasing number of patients requiring surgical treatment [222324]. The microendoscopic decompression technique for lumbar disorders involves bilateral decompression surgery using the unilateral approach. The procedure has been gradually and successfully applied to the treatment of lumbar spinal stenosis [2526]. We used the same technique for two portal percutaneous endoscopic decompression for lumbar spinal stenosis. This procedure has three advantages. First, ordinary arthroscopic and spine instruments are used, without the need of special endoscopic sets. Second is the free movement and angulation of the surgical tool and the endoscope independent of each other as they are not restricted by the confines of a common working portal, which results in marked reduction in technical difficulties. Third, the use of saline irrigation abolishes the problem of repeated cleaning of the endoscopic lens to remove accumulated fog or blood.

However, the surgical procedure has not yet been established internationally. One of the main barriers is the endoscopic approach to the spine. The spine is not a hollow space like the knee or shoulder, so arthroscopic visualization of the working instrument and target tissues/bone may not be easy, even with water pumping and irrigation. A potential space needs to be created before laminectomy (Fig. 2). Another hindrance is that the surgical technique involves partial resection of the bilateral lamina and total removal of the ligamentum flavum from the base of the spinous process using a pneumatic burr and rongeur. The procedure decreases perioperative complications and increases postoperative stability. Moreover, it is possible to completely preserve the contralateral facet joint. Previous investigation on facet joint resection rate after microendoscopic decompression surgery for lumbar spinal stenosis found that about 70% of the facet joints were preserved in the side of the approach and more than 95% were preserved in the contralateral side. However, maximal preservation of the facet joints is important for preventing the progression of spinal instability after surgery [1427].

In the present study, the most commonly encountered surgery-related complication was transient paresthesia of the nerve root, which has a previously reported incidence of 10% in open laminectomy for lumbar spinal stenosis. However, a learning curve is a reality of the technique concerning the surgical approach and severity of stenosis; most cases occurred during the early development of the technique, in patients undergoing unilateral laminotomy for bilateral decompression or in patients with severe stenosis. As shown in this study, percutaneous endoscopy results in an excellent viewing angle with good contralateral visualization. Hence, even in patients with severe stenosis, a dorsal space was first made by the bony procedure, so intra-operative positive pressure irrigation and the spinal decompression could be carried out more easily. Dural tearing does not occur, since the ligamentum flavum is kept as protective barrier for the dura mater until completion of the bony procedure. Manipulation of the neural tissues and extensive surgical skill of the surgeon may alleviate the risk of such complications.

There are some limitations associated with the current study. First, this study is retrospective. Second, the follow-up period was not long enough to show the long-term outcome. Because the initial benefits of surgical decompression may decline over time, additional longer-term studies should pay more attention to the risks of postoperative instability and restenosis as well as the need for re-operation.

Surgical decompression with two portal percutaneous endoscopic decompression has initial benefits, but long-term studies should pay more attention to the risks of postoperative instability and restenosis as well as the need for re-operation. Further investigations with long-term results are thus required.