European Journal of Orthopaedic Surgery & Traumatology

, Volume 24, Supplement 1, pp 231–237

The efficacy of epiduroscopic neural decompression with Ho:YAG laser ablation in lumbar spinal stenosis

Authors

  • Gun Woo Lee
    • Department of Orthopaedic SurgeryArmed Forces Yangju Hospital
  • Soo-Jin Jang
    • Department of Orthopaedic SurgeryKosin University Gospel Hospital
    • Department of Orthopaedic SurgeryKosin University Gospel Hospital
Original Article

DOI: 10.1007/s00590-013-1407-7

Cite this article as:
Lee, G.W., Jang, S. & Kim, J. Eur J Orthop Surg Traumatol (2014) 24: 231. doi:10.1007/s00590-013-1407-7

Abstract

Although epiduroscopy is one of the popular interventions for the management of lumbar spinal stenosis (LSS), only a part of these patients show improvement in pain and functional level. Consequently, the authors thought that holmium:YAG (Ho:YAG) laser can be a reasonable alternative as an adjunct of epiduroscopic procedure, but has not been thoroughly determined yet which influence is resulted by it. This study was conducted to evaluate and compare the efficacy of epiduroscopic neural decompression (END) and END with Ho:YAG laser (ELND) in patients with LSS. Forty-seven patients with LSS were enrolled, all of whom underwent END or ELND and were followed up for 2 years or more. Clinical outcomes were evaluated using the visual analog scale (VAS) for back and leg pain and the Roland Morris Disability Questionnaire (RMDQ). Procedure-related complications, especially including laser-related complications, were also evaluated. The only laser-related complication that occurred was transient mild motor paralysis in one case (3.1 %). In the END group, clinical score is exhibiting V-shaped upward trend that ended after procedure with the almost similar score obtained with preoperative status. However, in the ELND group, it is exhibiting relatively consistent improvement after procedure. There was a statistically significant improvement in the VAS and RMDQ score after 6 months after ELND procedure compared with END procedure (p = 0.01, 0.03, respectively). ELND could produce significant improvement of low back pain (LBP) at the last follow-up time (p = 0.01), but radiating pain of leg could not be improved significantly (p = 0.09). In conclusion, the current study suggests that performing Ho:YAG laser ablation concurrently with END could produce more decreased intensity of pain and prolonged effect of pain relief compared with END in LSS patients. LSS patients with LBP would be an ideal candidate for ELND, but radiating pain of LSS might not be managed effectively with ELND.

Keywords

EpiduroscopyHo:YAG laserLow back painLumbar spineSpinal stenosis

Introduction

Lumbar spinal stenosis (LSS) is a common disease that causes low back pain (LBP) and a variety of symptoms in the legs. Most patients are treated conservatively; however, conservative treatments such as medication and several injection techniques often provide only temporary relief and have limited effectiveness. Although surgical treatment is recommended for patients who do not respond to conservative treatments, symptomatic LSS frequently occurs in older adults and whose physical status is not always suitable for surgery [1].

Epiduroscopy is a new minimally invasive technique for the management of spine-originating problem. At the present time, a few studies have investigated the benefits of epiduroscopy for LSS, although these studies have mostly described both the clinical effectiveness and cost-effectiveness of epiduroscopy with retrospective design and small sample size [26]. Basically, epiduroscopic procedures offer potential advantages for the management of chronic LBP and/or leg pain in LSS owing to direct visualization of spinal structures, which allows focused adhesiolysis [5], dilusion of biomechanical irrigators such as phospholipase A2 (PLA2) and tumor necrosis factor (TNF)-α by saline irrigation [79], and targeted deposition of the injection [4]. Several clinical studies have also demonstrated that this novel technique can reduce pain and improve functional status through several therapeutic pathways as described above [2, 10, 11]. Based on its effectiveness, epiduroscopy for the management of spine-related problems is one of the most commonly performed interventions, but only a moderate proportion of patients show improvement in pain and functional level after epiduroscopic procedures. These relatively unsatisfactory outcomes suggest that additional methods should be sought to improve the efficacy of epiduroscopic procedures.

The authors had found the additive therapeutic methods simultaneously adapting with conventional epiduroscopic procedure for improving the clinical outcomes. Clinical adaptation of lasers has been shown to lead to favorable outcomes in other orthopedic territory [12]. On the spine territory, it has been demonstrated that laser can reduce pressure by vaporizing a certain volume of disc material, which reduces the pressure between the nucleus pulposus and the peridiscal tissue, causing retraction of the herniation away from the dura and nerve root [13]. With regard to orthopedics, laser therapy has been widely used to reduce the pain and disability associated with musculoskeletal and joint diseases of the shoulder and knee, such as rotator cuff disorders, subacromial impingement, and patellar impingement [14, 15]. In addition, the clinical trials show positive results in other regions [16, 17]. However, utilizing the laser simultaneous with epiduroscopic procedure for the management of LSS has not been thoroughly described. So far, various types of laser for managing spinal disorders have been reported at the clinical or experimental level, including the Nd:YAG laser [18, 19], frequency-doubled neodymium:yttrium–aluminum–garnet (Nd:YAG) laser [20], holmium:YAG (Ho:YAG) laser [21, 22], and semiconductor laser [23]. Of them, the Ho:YAG laser has a wavelength of 2.1 μm and reaches a depth in the tissue of ≤0.5 mm [24]. Tissue permeability is low, tissue vaporization is excellent, and the damage of surrounding tissue is minimal [24]. Consequently, the Ho:YAG laser may cause less damage to the surrounding tissues such as dura or nerve root in small enclosed spaces compared with other types of lasers.

The aims of this study were to evaluate the efficacy and drawback of epiduroscopic neural decompression with the Ho:YAG laser (ELND) in patients with LSS and to compare the clinical outcomes of ELND to those of epiduroscopic neural decompression (END) for a follow-up period of at least 2 years. To our knowledge, this is the first study to specifically address the benefits and drawbacks of ELND for a follow-up period of at least 2 years.

Materials and methods

Study design

We obtained approval from the institutional review board (IRB) for this study. All patients had initially presented with the symptoms of LSS, including neurogenic claudication and LBP with or without radiculopathy, and did not respond to conventional therapy, including medication, physiotherapy, or injection treatment such as epidural or caudal block, and selective nerve root block according to patient’s complaint. Radiologic workup for the preoperative evaluation included plain film radiographs, computed tomography (CT), and magnetic resonance images (MRI) for all patients.

Inclusion and exclusion criteria

This was retrospectively comparative study regarding postprocedure outcomes following END and ELND. Between October 2008 and December 2012, a total of 167 patients underwent END or ELND. Actually, our intention had been report on the outcomes of END from October 2008. However, in July 2009, the corresponding author altered technique to ELND because we had confidence regarding safety and efficacy of additive laser ablation by reviewing and introducing Ho:YAG laser. All operations were performed by one surgeon (the corresponding author).

Both procedures, END and ELND, were performed based on the following indications: (1) obvious findings of central canal stenosis concurrently with or without foraminal stenosis due to hypertrophied ligamentum flavum, facet hypertrophy, and bony compression such as bony spur on CT and MRI of lumbar spine; (2) obvious clinical presentation due to LSS compatible with radiologic findings in CT and MRI; (3) no improvement in LBP despite several conservative treatments for a minimum period of 3 months, where the conservative treatment included medication, physical therapy, and injection treatment. Inclusion criteria for this study were as follows: (1) patients for whom conservative treatment had failed for 3 months and (2) a follow-up period of 2 years or more after surgery. Exclusion criteria were as follows: (1) Patient follow-up was limited to 2 years or less; (2) failed back surgery syndrome; (3) patient not to be capable of diagnosing with LSS such as just lumbar disk herniation or spondylolysis without stenotic lesion; and (4) patient inability to accurately record results of preoperative and postoperative questionnaires due to problems such as a history of stroke, dementia, or major medical illness requiring intensive treatment. These inclusion/exclusion criteria were used to limit the number of confounding variables and to make it easier to determine the effects of additive laser ablation. Prior to surgery, all patients were informed regarding the details of their operations, including type, timing, difficulty, and potential complications. An informed consent was obtained from all patients.

Epiduroscopic procedures with or without laser ablation

Each patient was instructed to lie on a radiological testing table in the prone position; the skin in the region of the sacral hiatus was swabbed with disinfectant for sterilization, and local anesthesia was administered. Next, a 1-cm-long longitudinal incision was made on the skin in the sacral hiatus region; an 18G Tuohy epidural needle was used to puncture the sacral hiatus, and a guide wire was inserted into the opening via the needle that was removed. The introducer was then inserted followed by a flexible endoscope (Myelotec, model # 75298871, Roswell, GA, USA), which was inserted into the anterior and posterior epidural space through the introducer. The flexible endoscope was maneuvered to reach the suspected lesion. Epidural saline solution was used to irrigate and clear the area visualized, to exert pressure for expansion of the epidural space, and to improve clarity on the endoscopic video screen. This allowed visualization of adhesion status with bands, inflammatory tissues, fibrous connective tissues, and adipose tissue around the dura and nerve root, as well as anterior and posterior stenotic lesions such as thickened ligament flavum, hypertrophied facet joint, and bulged intervertebral disk. In addition to securing epidural visibility epiduroscopically, we injected a contrast agent (Omnipaque™, Nycomed, Ireland) in order to perform epidurography that identifies the adhesive lesion clearly.

In the group of patients who underwent END with Ho:YAG laser ablation (ELND), we performed laser-based cauterization and decompression of the disk after adhesiolysis and foraminoplasty. Before we performed decompression and adhesiolysis, we introduced a guide wire probe under endoscopic visualization directly into the causative factors via the working channel. When the anterior and posterior causes were cauterized via the endoscopic camera, the causes were shrunk, and dura and nerve roots would be decompressed. Both groups received a locally injected steroid, triamcinolone 80 mg (Kena-log®, Bristol Myers Squibb Co. Princeton, NJ, USA), directly in the treatment area to maximize the treatment effects.

During procedure, no other sedatives were administered except the intravenous injection of meperidine hydrochloride (Demerol® 25 mg, Sanofi-aventis U.S. LLC, Bridgewater, NJ, USA) to help ease the patients’ pain occurred during procedure and to facilitate verbal communication between the surgical staff and patients throughout the procedure.

Data collection

Demographic data and clinical data were prospectively recorded. In addition, subjects completed a questionnaire, consisting of a 10-point visual analog scale (VAS) for LBP and leg pain (in which 0 indicates no pain and 10 the worst pain possible) preoperatively and at each follow-up visit; subjects also completed the Roland Morris Disability Questionnaire (RMDQ) at each follow-up visit. The questionnaire was completed by patients before they met with the surgeon at each follow-up visit and was submitted to an assistant who tabulated and scanned the forms. Follow-up visits typically occurred at 1, 3, 6, 12, and 24 months after surgical procedure. And procedure-related complications, especially including laser-related complications, were also evaluated.

Statistical analysis

Continuous variables were compared with repeated-measures ANOVA, and categorical variables were compared with Fisher’s exact tests and Student’s t tests. We used SPSS software version 18.0 (SPSS, Chicago, IL, USA) for the statistical analysis. A p value <0.05 was considered statistically significant.

Results

Patient population

Of total 167 patients, 57 patients, twenty-five consecutive patients (11 men and 14 women) in the END group and 32 consecutive patients (15 men and 17 women) in the ELND group, were enrolled in this study and evaluated (Fig. 1). Demographic data of the two groups are summarized and compared in Table 1. The mean age was 67.1 ± 7.8 and 65.2 ± 6.4 years in END and ELND groups, respectively. The mean follow-up period was 34 ± 3.1 months for the END group and 28 ± 2.5 months for the ELND group. There was no significant difference in age (p = 0.12), gender (p = 0.09), and follow-up period (p = 0.07) between two groups (Table 1).
https://static-content.springer.com/image/art%3A10.1007%2Fs00590-013-1407-7/MediaObjects/590_2013_1407_Fig1_HTML.gif
Fig. 1

Flow diagram for patients’ enrollment

Table 1

Comparison of the demographic data between both groups

 

END

ELND

p*

N

25

32

0.07

Age (years)

67.1 ± 7.8

65.2 ± 6.4

0.12

Male/female

11:14

15:17

0.09

Follow-up periods (months)

34 ± 3.1

28 ± 2.5

0.07

END epiduroscopic neural decompression, ELND epiduroscopic laser neural decompression

Clinical outcomes

In the END group, the mean VAS score for LBP was 8.5 ± 1.3 preoperatively, 3.8 ± 0.4 at 1 month after surgery, and 6.7 ± 0.9 at 2 years after surgery (p = 0.12), exhibiting a “V-shaped” upward tendency, and the mean VAS score for leg pain was 6.7 ± 0.8 preoperatively, 3.1 ± 0.7 at 1 month after surgery, and 5.2 ± 0.6 at 2 years after surgery (p = 0.15). The mean RMDQ score was 12.6 ± 1.2 preoperatively, 8.6 ± 0.6 at 1 month after surgery, and 10.4 ± 0.5 at 2 years after surgery (p = 0.09), also exhibiting V-shaped upward trend in which the score at final follow-up was almost identical to that before surgical treatment (Table 2, 3).
Table 2

Summarized data of clinical outcomes

 

END (N = 20)

ELND (N = 27)

p

VAS for low back pain

   

 Preprocedure

8.5 ± 1.3

8.1 ± 0.7

0.41

 1 month later

3.8 ± 0.4

3.1 ± 0.3

0.11

 6 months later

4.1 ± 0.5

3.0 ± 0.7

0.03*

 12 months later

4.9 ± 0.3

3.5 ± 0.6

0.02*

 24 months later

6.7 ± 0.9

4.4 ± 0.5

<0.01*

VAS for leg pain

   

 Preprocedure

6.7 ± 0.8

6.2 ± 0.9

0.27

 1 month later

3.1 ± 0.7

2.8 ± 0.4

0.06

 6 months later

3.2 ± 0.4

2.9 ± 0.3

0.07

 12 months later

4.5 ± 0.8

3.9 ± 0.4

0.03*

 24 months later

5.2 ± 0.6

4.7 ± 0.6

0.05

RMDQ

   

 Preprocedure

12.6 ± 1.2

13.2 ± 0.9

0.32

 1 month later

8.6 ± 0.6

7.2 ± 0.5

0.04*

 6 months later

8.5 ± 0.9

6.8 ± 0.5

0.01*

 12 months later

9.7 ± 1.1

7.1 ± 0.6

<0.01*

 24 months later

10.4 ± 0.5

8.5 ± 0.3

<0.01*

END epiduroscopic neural decompression, ELND epiduroscopic laser neural decompression, VAS visual analog scale, RMDQ Roland Morris Disability Questionnaire

* Statistically significant (p < 0.05)

Table 3

Clinical outcomes in both groups

 

Preoperative data

Last follow-up data

p*

END

   

 VAS for LBP

8.5 ± 1.3

6.7 ± 0.9

0.12

 VAS for RP

6.7 ± 0.8

5.2 ± 0.6

0.15

 RMDQ

12.6 ± 1.2

10.4 ± 0.5

0.09

ELND

   

 VAS for LBP

8.1 ± 0.7

4.4 ± 0.5

0.01

 VAS for RP

6.2 ± 0.9

4.7 ± 0.6

0.07

 RMDQ

13.2 ± 0.9

8.5 ± 0.3

0.03

END epiduroscopic neural decompression, ELND epiduroscopic laser neural decompression, VAS visual analog scale, LBP low back pain, RP radiating pain of leg, RMDQ Roland Morris Disability Questionnaire

* Statistically significant (p < 0.05)

In the ELND group, the mean VAS score for LBP was 8.1 ± 0.7 preoperatively, 3.1 ± 0.3 at 1 month after surgery, and 4.4 ± 0.5 at 2 years after surgery (p = 0.01), demonstrating a relatively consistent improvement during follow-up, and the mean VAS score for leg pain was 6.2 ± 0.9 preoperatively, 2.8 ± 0.4 at 1 month after surgery, and 4.7 ± 0.6 at 2 years after surgery (p = 0.07). The mean RMDQ score was 13.2 ± 1.6 preoperatively, 7.2 ± 0.5 at 1 month after surgery, and 8.1 ± 0.3 at 2 years after surgery (p = 0.03), also demonstrating a relatively consistent improvement during follow-up with regard to functional activity. There was a statistically significant improvement in the VAS and RMDQ score after 6 months after ELND procedure compared with END procedure (Table 2, 3).

Complications

We experienced several epiduroscopy-related complications, including transient headache (N = 3), pain over the site of the scope insertion (N = 3), focal infection of the entry site (N = 2), and mild meningitis (N = 1). However, all symptoms improved after bed rest and treatment with medications. The only laser-related complication that occurred was transient mild motor paralysis in one case (3.1 %); the patient recovered within 1 month after the procedure.

Discussion

This study compared the effectiveness of epiduroscopy with and without the use of a laser for LSS with LBP and/or radiating pain that was intractable to conservative treatments. We found that ELND offered superior therapeutic possibilities and nearly similar postprocedure complications as the END. In addition, our results suggest that ELND provides pain relief and functional recovery for a longer period of time than END.

Review of relevant literatures on the epiduroscopy with laser

Epiduroscopic procedure is a minimally invasive technique for direct visualization and therapeutic treatment of spinal disorders and can be performed under local anesthesia. Epiduroscopic procedures demonstrate superior effectiveness compared with both physical therapy [25] and caudal epidural injections [26]. However, recent systematic reviews indicate that this procedure is still controversial, with some studies showing positive results for short- and medium-term pain relief [2, 4, 5, 18, 26] and some studies showing negative results [27].

Given the conflicting and relatively unsatisfying results of epiduroscopy, we have found the additional therapeutic option during epiduroscopic procedure. In fact, the use of lasers in many applications remains controversial. For example, there is concern that the removal of tissue may be inadequate or uncontrolled. There is also concern that it may cause degradation of subchondral bone and thermal injury of important surrounding structures, such as the dura mater and nerve roots. However, laser has advanced rapidly over the past 10 years with the development of different lasers of varying wavelengths and a range of associated physical properties. The Ho:YAG laser appears to be the ideal wavelength and mostly absorbed by water in the tissue and promotes hemostasis, ablates tissue in a controlled and predictable fashion, and appears to cause minimal damage to adjacent tissues. We hypothesized that the Ho:YAG laser would be the most suitable form of laser for utilizing concurrently with END procedure in LSS. Additionally, the advancement of fiber-optic scopes allows for simultaneous laser ablation and cool saline irrigation, which reduces thermal damage to the nerves significantly.

Although there have been several contradictory reports on the advantages and disadvantages of END, we were unable to find, in a MEDLINE search, a comparative study with prospectively collected data that specifically investigating the outcomes of END with and without laser ablation. This study suggests that the combination of laser and epiduroscopy is an effective alternative in the management of LSS that is refractory to conservative treatments, such as medication or injection treatment. Several factors could be responsible for the observed pain relief and functional recovery after ELND. First, we administered local anesthetics and steroids at the end of epiduroscopic procedure. Local anesthetics induce sympathetic nerve blockade and therefore improve blood flow to the ischemic neural elements. Steroids reduce inflammatory edema of the injured nerve root and thus improve intraneural blood flow. Although the efficacy of epidural steroids in patients with spinal stenosis is controversial, the lack of positive effects in clinical studies may in part reflect incorrect placement of the injection. However, in epiduroscopy, exact placement of therapeutic agents such as steroid and analgesics can be achieved under a direct visualization of pathologic lesions. We think that initial improvement of symptoms after END or ELND may reflect local anesthetics and steroids that have reached the level causing symptoms. Secondly, we distended the epidural space by administering saline during epiduroscopy. Several studies have described the importance of inflammatory mediators such as PLA2 and TNF-α in causing low back and leg pain [79]. We suggest that pain relief in these patients may also be related in part to the effects of saline in diluting local concentrations of inflammatory mediators. Third, lysis of perineural adhesions during epiduroscopy may play a role in the lasting relief of pain. For our patients, we observed adhesions and fibrosis in the epidural space, which can result from compression, entrapment, and inflammation around the involved neural tissues. Fibrosis causes leg pain by interfering with the mobility of the dural sleeves of the spinal roots. It is possible that the mobility of the nerve roots is restored to some extent after epiduroscopy, and this may contribute to long-term pain relief that exceeds the short-term effectiveness of epidural injection. Lysis of epidural adhesions is a controversial topic in the treatment of patients with failed back surgery. Several researchers have stressed that lysis of epidural adhesions is an effective therapeutic way, whereas others have asserted that there is no correlation between therapeutic effect and the degree of lysis of epidural adhesions. Nevertheless, lysis of adhesions may contribute to the correct placement of injectate in the involved region of spinal stenosis, since epiduroscopy revealed various amounts of fibrous tissue and adhesions at the involved vertebral levels. Finally and perhaps most importantly, the use of a laser reduces mechanical compression of the cauda equine, nerve root, and DRG. The laser fiber can be introduced through the working canal of the epiduroscope. Under epiduroscopic visualization, the laser enables precise removal of morphological pathologic structures. Furthermore, hemorrhage and the extraction of tissue segments can be avoided [28].

Therapeutic efficacy of END and ELND

Catheter-induced interventional treatments such as epiduroscopic procedure have been introduced for clinical manifestations occurred due to LSS, but there has not been determined the best treatment modality yet. In the current study, END could not produce optimal clinical outcomes following procedure in terms of pain intensity (VAS) and functional status (RMDQ). Actually, this disappointing outcome had been also reported in some previous articles. To improve deficient therapeutic efficacy of END, the authors considered the epiduroscopic procedure, as well as performing additive laser ablation with epiduroscopic procedure. Based on the current study, ELND would produce greater clinical outcomes compared with END, and ENLD could be a reasonable alternative for LSS with LBP, but might not be effective for LSS with radiating pain. In other words, epiduroscopic procedure, END or ELND, had a limitation that radiating pain on lower extremity could not be managed effectively with it. This outcome was disappointed, but several therapeutic modalities such as selective nerve root block have been described for radiating pain, with favorable outcomes [29]. Thus, ELND would guarantee the achievement of great clinical improvement in LSS patients with LBP, but in radiating pain of LSS, ENLD should be performed concurrently with a therapeutic modality for managing radiating pain, including selective nerve root block and epidural block with several approaching techniques, and further studies should be addressed for evaluating therapeutic efficacy.

Limitations

As with any study, there are a number of limitations with ours. First of all, although we performed comprehensive analysis of all data, this is nonetheless limited by not randomized and prospectively designed study. Because these cases were not randomized and the ELND group was done at a later time, there are potential problems. However, we know of no way that one could adversely influence the outcomes of the END procedure or improve the outcomes of the ELND group. Second, further studies with larger sample size and longer follow-up period would also address the issue of the relative small sample size and short period included in this study.

Despite all the limitations, we believe that this study has unique strength. We performed a comparative analysis with relatively large numbers of patients in both groups. Considering that there have not been any studies regarding comparison of epiduroscopic procedure with or without laser for a follow-up of 2 years or more, this study should provide a foundation for further literatures.

Conclusion

To our knowledge, this is the first comparative study to investigate the outcomes of patients undergoing END with and without Ho:YAG laser ablation for 2 years or more in LSS. Our results demonstrated that ELND could achieve significant improvement of LBP in LSS at 2-year follow-up time, but have a limitation for improvement of radiating pain in LSS. And, ELND could produce significantly greater clinical improvements compared with END. On the basis of these results, we recommended that ELND could achieve great clinical improvement in LSS patients with LBP.

Conflict of interest

The authors have no personal financial or institutional interest in any of the drugs, materials, or devices described in this article.

Copyright information

© Springer-Verlag France 2014