Introduction

Degenerative lumbar foraminal stenosis is a condition in which a nerve root or spinal nerve is entrapped in a narrowed lumbar foramen in degenerative lumbar spinal disorders [9, 13, 20]. The circumferential narrowing of the space available for the exiting nerve root leads to back pain and radicular symptoms. Impingement of the foraminal canal has also been implicated as a cause of failed back surgery syndrome due to an unrecognized or possibly inadequate treatment of foraminal stenosis [5]. Therefore, it is an important pathology in recognizing outcomes both clinically and radiographically in the treatment of patients with radicular pain. However, the preferred surgical treatment to relieve the compression on the exiting nerve root has not been established yet. Several different techniques for degenerative lumbar foraminal stenosis have been described, including foraminotomy, facetectomy, partial pediculectomy, fusion, and distraction instrumentation. Traditionally, decompression with a spinal fusion is utilized as a primary treatment in patients with degenerative lumbar foraminal stenosis, even in the absence of a gross instability, partly because of the limitations of the effective decompression tools. To solve these problems, several authors reported that a paraspinal approach [21, 27] and various minimally invasive techniques have been developed [15, 22, 24, 29]. We performed endoscopic partial facetectomy (EPF) using an endoscopic osteotome for degenerative lumbar foraminal stenosis. The purpose of this study is to describe the EPF procedure and report its clinical and radiological outcomes.

Materials and methods

Between January 2012 and January 2014, EPF was performed on 32 patients. The patients underwent relevant clinical and radiological investigations, and an adequate follow-up was possible in 25 patients. Seven patients were excluded owing to an incomplete follow-up. One patient was lost to follow-up because she had undergone a revision fusion surgery in another hospital due to severe radicular symptoms by our phone request information. Six patients were also lost to follow-up because they did not come to the hospital despite our phone requests for the postoperative follow-up. A total of 25 consecutive patients (12 women and 13 men) who underwent EPF were retrospectively reviewed. The inclusion criteria of this study were degenerative lumbar foraminal stenosis with unilateral radicular symptoms. Patients with clear instabilities, central canal stenosis, and/or bilateral symptoms were excluded. All patients underwent a continuous conservative treatment, including medicine, physical therapy, and selective nerve root block therapy for a minimum of 6 weeks before surgery. The radiographic indications were moderate to severe foraminal stenosis with perineural fat obliteration or nerve root collapse based on the Wildermuth grading system [26]. All patients were assessed before surgery and followed-up regularly during outpatient visits (preoperatively and 1, 3, 6, 12, and 24 months postoperatively). The study was approved by the Clinical Research Ethics Committee of Pusan National University Hospital.

Surgical technique

The EPF procedure was performed under local anesthesia. As a premedication, midazolam (0.05 mg/kg) was injected intramuscularly 30 min before surgery. Dexmedetomidine (1 µg/kg within 10 min for the loading dose and 0.2–0.7 µg/kg/h for the maintenance dose) was intravenously administered during surgery. The patients were placed in the prone position on a radiolucent table. The skin entry point was located on the lateral edge of the paravertebral back muscle (about 8–13 cm lateral to the midline, depending on the patient’s waist size), and local anesthesia was administered to the intervertebral foramen and facet joint. The target point of initial needling was the surface of the facet joint, with the needle firmly engaged into the facet joint and then replaced by a guidewire under a lateral fluoroscopic control. A tapered obturator was inserted over the guidewire not to the intervertebral foramen but to the facet joint, which prevents an exiting nerve root damage and enables a safe working space without a direct contact on the exiting nerve root (Fig. 1). After the correct placement of the obturator on the facet surface, a beveled working cannula was introduced over the obturator. After the obturator was withdrawn, an endoscope was inserted. Initially, the surgeon can view the surface of the facet joint via an endoscopic visualization. After the soft tissue, capsule, and cortical bone of the facet joint were removed using an endoscopic burr and a low-energy bipolar radiofrequency, the cancellous bone of the facet joint was exposed (Fig. 2a and b). Once the joint was identified via endoscopic visualization, osteotomy of the superior articular process was performed using several custom-made endoscopic osteotomes (Fig. 3a and b). Although the superior articular process can be mobilized after osteotomy, removal of the bony fragments may be difficult. Because the capsule and ligament were firmly attached to the facet, the bony fragments can be removed by manipulating with a beveled obturator engaged to the osteotomy site (Fig. 3c and d). After the hypertrophied superior articular process was removed, the intraforaminal structures, such as the foraminal ligament, ligament flavum, perineural fat, and disc surface should appear clearly. Thereafter, the hypertrophied ligament flavum and osteophyte around the exiting nerve root were removed using endoscopic forceps and punch. A flexible probe under endoscopic view was used to confirm an adequate decompression by mobilizing the exiting nerve root (Fig. 4a and b). After all instruments were removed, direct closure of the skin was performed. No drainage was required. All patients were compared in terms of the extent of decompression via magnetic resonance imaging preoperatively and 1 day postoperatively (Fig. 5a and b), discharged within 5 days after surgery, and provided with a back brace for 2 weeks.

Fig. 1
figure 1

A tapered obturator is inserted over the guidewire to the L5/S1 facet joint under a fluoroscopic lateral view. The discography is performed routinely

Full size image
Fig. 2
figure 2

a The capsule and cortical bone of the facet joint are removed using an endoscopic burr. b The cancellous bone of the facet joint is exposed after decortications using an endoscopic burr

Full size image
Fig. 3
figure 3

a Once the joint is identified via endoscopic visualization, osteotomy of the superior articular process is performed using an endoscopic osteotome. b Osteotomized superior articular process on an intraoperative endoscopic view. c Because the capsule and ligament are firmly attached to the facet, the bony fragments can be removed by manipulating with a beveled obturator engaged to the osteotomy site. d After the hypertrophied superior articular process is removed, the articular cartilage in the facet joint is seen on an intraoperative endoscopic view

Full size image
Fig. 4
figure 4

a and b A flexible probe under endoscopic view can be used to confirm the decompression by mobilizing the exiting nerve root

Full size image
Fig. 5
figure 5

Illustrated case of a 70-year-old male patient. a Preoperative magnetic resonance (MR) image showing foraminal stenosis with disc herniation at the right L5/S1 level. b Postoperative MR image showing decompression of the stenotic foramen and visualization of the exiting nerve root 1 day after surgery

Full size image

Measurements of the clinical and radiological outcomes

At each visit, the patients were asked to complete an outcome questionnaire, and plain radiographs were obtained. The clinical outcomes were evaluated using the visual analog scale (VAS), Oswestry Disability Index (ODI), and Short Form-36 (SF-36) outcome questionnaire (preoperatively and 1, 3, 6, 12, and 24 months postoperatively). The radiological parameters were measured preoperatively and 1 and 2 years postoperatively. The radiological outcome was measured using the lumbar Cobb angle, disc wedging angle, lumbar lordosis (LL), slip percentage, and disc height index (DHI) in plain standing radiographs. Plain standing radiographs were obtained in all patients, and changes in the radiographs were evaluated in terms of the lumbar Cobb angle and disc wedging angle at the affected level on the anteroposterior view. The LL, slip percentage, and DHI were checked at the affected level on the lateral plain radiograph. The slip percentage, which is the degree of slip as a percentage of the anteroposterior diameter of the superior margin of the lower vertebra, was calculated using the Taillard method [25]. The DHI, which is the ratio of the mean of the disc height (three-point disc height at each operated level) to the mean sagittal diameter of the consecutive vertebra, was calculated from the midvertebral level using the modified Inoue method [11] (Fig. 6). Radiological progression was defined as ≥5° Cobb angle progression, ≥3° disc wedging progression, ≥5% slip percentage progression, and/or ≥2 mm disc height (≥5% DHI) progression [29]. All measurements were performed twice independently by three spine surgeons with an interval of 2 weeks between the two measurements to decrease intraobserver (Pearson correlation coefficient, 0.919; range, 0.884–0.940) and interobserver errors (Pearson correlation coefficient, 0.910; range, 0.873–0.930).

Fig. 6
figure 6

Radiologic measurement of DHI at the affected level. A, B (white line): sagittal diameter from the midvertebral level of the consecutive vertebra; 1 (black line): the line between the anterior/superior corner of the upper vertebra and anterior/inferior corner of the lower vertebra; 2 (black line): the line between the middle/superior point of the upper vertebra and middle/inferior point of the lower vertebra; 3 (black line): the line between the posterior/superior corner of the upper vertebra and posterior/inferior corner of the lower vertebra; a, b, c measured disc height on lines 1, 2, and 3. Disc height index (DHI) measured using the modified Inoue method was calculated using the following formula: DHI (%) = {[(a + b + c)/3]/[(A + B)/2]} × 100

Full size image

Statistical analysis was performed using the SPSS 21 software for Windows (SPSS, Chicago, IL, USA). Data were expressed as means ± standard deviations. The clinical and radiological outcomes were compared using the ANOVA test and t-test, if appropriate. The post-test was used to analyze individual differences. A P < 0.05 was regarded as statistically significant.

Results

Adequate clinical and radiological follow-ups were available for 25 patients. The average age at the time of surgery was 66.1 (48–82) years. Twelve patients were women, and 13 patients were men. Four patients had surgery at two levels (L3/4/5 in one case and L4/5/S1 in three cases), and 21 patients had surgery at a single level (L2/3 in two cases, L3/4 in one case, L4/5 in nine cases, and L5/S1 in nine cases). Of the EPF performed, 15 cases were left-sided, and 10 cases were right-sided. We performed EPF with an additional discectomy (EPF with discectomy group) in 11 patients and only EPF without discectomy (EPF group) in 14 patients.

The clinical outcomes were evaluated using the VAS, ODI, and SF-36 outcome questionnaire (preoperatively and 1, 3, 6, 12, and 24 months postoperatively). The baseline preoperative mean values of the VAS, ODI, and SF-36 were 74.6, 48.1, and 34.4, respectively. Compared with the baseline mean values, the 1- and 3-month follow-up values showed significant improvements. Similar results were observed for the 2-year follow-up scores (Table 1). The EPF with discectomy group and the EPF group were not significantly different in terms of VAS, ODI, and SF-36 scores (Table 2).

Table 1 Clinical outcomes
Full size table
Table 2 Clinical outcomes between EPF with discectomy group and EPF group
Full size table

The radiological parameters were measured preoperatively, and 1 and 2 years postoperatively (Table 3). There were no significant differences in the lumbar Cobb angle, disc wedging angle, LL, and slip percentage preoperatively, and 1 and 2 years postoperatively. The DHIs decreased during the follow-up periods compared with the preoperative baseline values; however, there was no statistical significance. In addition, the preoperative and 1- and 2-year postoperative DHIs did not significantly differ between the EPF with discectomy group and the EPF group (Table 4).

Table 3 Radiological outcomes
Full size table
Table 4 Radiological outcomes between EPF with discectomy group and EPF group
Full size table

In the present study, there were four patients with transient postoperative dysesthesia, which improved during the follow-up with medications and selective nerve root block. There were no other serious complications in our study. Follow-up images on dynamic X-ray showed no instability at the operated or any adjacent segments.

Discussion

Loss of intervertebral disc height secondary to degeneration allows the superior articular process of the inferior vertebra to subluxate anteriorly and superiorly, diminishing the foraminal area. Lumbar exiting nerve roots may be compressed in anteroposterior and/or craniocaudal directions [13, 19]. Anteroposterior foraminal stenosis attributes to transverse compression between the superior articular facet and posterior vertebral body. Craniocaudal foraminal stenosis is caused by a protrusion of the posterolateral osteophytes from the vertebral endplates into the foramen with a laterally bulging disc, compressing the nerve root against the superior pedicle [29]. The L5 nerve root may also be compressed by the L5 transverse process, sacral ala, or lumbosacral ligament in the extraforaminal zone [18, 28]. Moreover, dynamic factors contribute to foraminal stenosis [7, 12].

Two surgical treatments are available for patients with degenerative lumbar foraminal stenosis: total facetectomy with/without fusion and facet-preserving decompression. Total facetectomy offers sufficient decompression through the nerve root course. However, this often leads to segmental instability and back pain, which may eventually necessitate revision surgeries with fusion [1, 6, 8, 13, 21]. Important structural ligaments that can be affected with conventional facetectomy include the supraspinous, interspinous, and intertransverse ligaments with the superficial thoracolumbar fascia and various interspinous muscles depending on the approach and technique used. Although fusion is usually indicated for preoperative instabilities or deformities, facet-preserving decompression is more desirable as the primary treatment in patients with degenerative lumbar foraminal stenosis in the absence of gross instabilities. Haufe and Mork [10] reported good results in the sagittal rotational or translational motion after unilateral endoscopic total facetectomy for the treatment of severe foraminal stenosis. However, their total facetectomy using the endoscopic system can cause instabilities of the spine. It is well known that the facet joints have an important role in stabilizing the lumbar joints, and the posterior annulus is protected by the facet joint, especially in axial rotation [2, 16, 17]. So a facet-preserving foraminotomy through a paraspinal approach was introduced [21, 27], and many surgeons reported good results using modified methods of this technique [4, 8, 10, 23, 29].

Recently, endoscopic lumbar foraminotomy has been performed, but there is a concern about safety because an endoscopic burr should be used to decompress the stenotic foramen [4]. We performed EPF using an osteotome, which is a more simplified and safe procedure than foraminoplasty using a burr, for degenerative lumbar foraminal stenosis. None of the patients had any statistically significant changes in the radiological parameters, including instability and progression of scoliosis or kyphosis after EPF for 2 years of follow-up. Thus, EPF can be one of the safe and sufficient methods for the treatment of degenerative lumbar foraminal stenosis in selected cases. An advantage of EPF is that it can be safely performed without the need for fusion or instrumentation and preserve the stability of the spine. However, as with other endoscopic surgeries, the inherent disadvantage of EPF is its high exposure to radiation. Ahn et al. [3] compared the radiation dose of percutaneous endoscopic lumbar discectomy (PELD) with open lumbar microdiscectomy and minimally invasive surgery (MIS) microdiscectomy. The mean radiation dose of PELD was much greater than that of MIS microdiscectomy or open microdiscectomy. This implies that a less invasive discectomy technique inevitably results in more radiation exposure for the surgeons, and medical staffs.

Recently, several minimally invasive techniques for the treatment of lumbar foraminal stenosis have been described in the literature with good results [4, 14, 29]. In the present study, the clinical outcomes of VAS, ODI, and SF-36 significantly improved after surgery. Furthermore, the clinical outcomes between the EPF with discectomy group and EPF group were not significantly different. Because an aggressive discectomy may cause an exiting nerve root injury during the EPF procedure and/or loss of disc height at follow-up, only EPF or EPF with a minimal discectomy, as necessary, will be required in patients with degenerative lumbar foraminal stenosis.

This study has several limitations that require consideration. First, the number of subjects tested was relatively small and no control group was used, which diminished the statistical power of the study. Second, because of the relatively short follow-up period, further studies with larger number of patients and long-term outcomes are needed. Third, even though the clinical and radiological outcomes were similar between the EPF with discectomy group and EPF group, we do not know the correlation between the degree of discectomy and outcomes because we did not check the volume, extent, and ratio of the removed disc contents. Fourth, seven patients were excluded owing to an incomplete follow-up. Fifth, we did not distinguish between back pain and leg pain, but overall pain was measured by VAS. Assessment of overall pain using VAS may not accurately reflect radicular problems.

In conclusion, EPF is a safe and effective option in decompressing the lumbar exiting nerve root without causing spinal instabilities for the treatment of patients with lumbar foraminal stenosis.