The role of epidural contrast distribution in predicting the effectiveness of steroid and local anesthetic injection in patients with lumbar spinal stenosis

Abstract

STUDY DESIGN:

A prospective observational comparative study.

BACKGROUND:

Because epidural corticosteroids are more effective for pain relief when delivered close to the site of pathology, lumbar spinal stenosis with multilevel stenotic areas can be treated more effectively when medications cover all of the stenotic areas. Distribution of medications to the pathologic sites is considered an important factor in the effectiveness of inter-laminar epidural injection.

OBJECTIVE:

To determine whether spread of a lidocaine and triamcinolone mixture over the stenotic areas improves clinical symptom of spinal stenosis more effectively.

METHODS:

Twenty-four patients with lumbar spinal stenosis were enrolled in this study. A lumbar inter-laminar epidural injection under fluoroscopic guidance was administered to each patient. A numeric rating scale (NRS), claudication distance, the Romberg test, and the Oswestry Disability Index (ODI) were checked pre-injection and at 2, 6, and 12 weeks after the epidural injection. Vertebrae that showed more than moderate stenosis were considered stenosis levels. Inter-laminar epidural injection was performed with radiopaque contrast and lidocaine mixed with triamcinolone acetate. When the contrast covered all of the stenosis levels, this was considered a sufficient distribution (SD) and when the contrast did not cover all stenosis levels, this was considered an insufficient distribution (ID).

RESULTS:

At 2 weeks, percentages of NRS improvement were 71.4 (IQR, 20.2) in SD group and 50.0 (IQR, 31.4) in ID group ( $P=$ 0.02), changes in Romberg test time were 0.0 (IQR, 35.0) in SD group and $-$ 3.0 (IQR, 12.0) in ID group ( $P=$ 0.02), decreases in ODI were 9.6 (IQR, 11.2) in SD group and 0.0 (IQR, 7.8) in ID group ( $P=$ 0.02). However, these parameters were not significantly different at 6 weeks and 12 weeks. Claudication distance showed no significant group difference.

CONCLUSION:

Coverage of the stenotic area by the lidocaine, triamcinolone, and contrast dye mixture during inter-laminar epidural injection was an important predictor of acute pain reduction.

Keywords

Spinal stenosis epidural injection contrast distribution inter-laminar

1. Introduction

Spinal stenosis involves narrowing of the spinal canal and causes symptoms of neurogenic claudication [1]. Many nonsurgical treatments have been evaluated, including medications, physiotherapy, and epidural steroid injection [2]. However, high- and moderate-grade evidence for these treatments is lacking [3] and no definitive clinical guidelines have been suggested to date.

Epidural injection is a procedure in which soluble agents, such as, local anesthetics or steroids are injected into the epidural space by transforaminal, inter-laminar, or caudal approaches to control pain. Among these, inter-laminar epidural injection is most commonly performed, and shows short term pain relief in cases of herniated cervical or lumbar disc, spinal stenosis, discogenic pain, and radiculopathy [4]. Because the inter-laminar approach spreads medications to multiple levels of the lumbar spine [5], it is a good choice in cases of diffuse targets, such as multilevel spinal stenosis [6].

The efficacy of inter-laminar epidural injection in lumbar spinal stenosis depends on a number of factors, including age and sex [7]. The severity of lumbar spinal stenosis has a debatable correlation with clinical improvement due to epidural injection [7, 8, 9]. Because epidural corticosteroids are more effective for pain relief when delivered close to the site of pathology [10], lumbar spinal stenosis with multilevel stenotic areas can be treated more effectively when medications cover all of the stenotic areas. In cases of lumbosacral radicular pain, parasagittal inter-laminar epidural injection is more effective for pain relief than midline inter-laminar epidural injection because the needle tip is closer to the pathologic lesions [11]. Therefore, distribution of medications to the pathologic sites is considered an important factor in the effectiveness of inter-laminar epidural injection.

We hypothesized that if the contrast dye covered the stenotic areas, inter-laminar epidural injection would be more effective in relieving pain associated with spinal stenosis. Therefore, the aim of this study was to determine whether coverage of the contrast dye over stenotic areas results in greater improvement of the clinical symptoms of spinal stenosis.

2. Methods

2.1 Patients

This was a prospective observational comparative study. Patients were recruited from our outpatient department who were diagnosed with spinal stenosis confirmed by magnetic resonance imaging (MRI) from March 2012 to November 2012. Inclusion criteria were 1) claudication symptoms and signs irrespective of low back pain, 2) no resolution of pain by medication or physical modalities, 3) central stenotic regions revealed by lumbosacral MRI, 4) neither hip joint nor vascular problem by physical examinations, 5) no previous intervention within 3 months before enrollment, 6) no previous lumbosacral surgery, and 7) no contraindication for the epidural injection, such as, pregnancy or acute infection. Epidural fibrosis has been reported in over 80% of patients after surgery [12]. Therefore, we excluded patients who underwent lumbosacral surgery.

All patients gave written consent for this procedure, and their information was acquired. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

2.2 Parameters

A numeric rating scale (NRS) ranging from 0 to 10, claudication distance, the Romberg test, and the Oswestry Disability Index (ODI) were measured pre-injection and at 2, 6, and 12 weeks after epidural injection. The version of the ODI was the same as that used in a previous study [13]. Percentage (%) of NRS improvement after the epidural injection was expressed as the changes of NRS divided by the initial NRS. Percentage (%) of NRS improvement at more than 70% was considered as ‘more effective in NRS’. Less than 70% NRS improvement was considered as ‘less effective in NRS’. A decrease of ODI of more than 8 or less than 8 was considered as ‘more effective in ODI’ or ‘less effective in ODI’, respectively. We determined the criteria of responder as 8 point according to a previous study in which median difference of ODI between ‘better’ group and ‘worse’ group was 8 [14]. Basic patient information was collected, including height and body weight. Claudication distance was measured on the treadmill with a usual walking speed ranged from 5 to 6 km per hour until the symptoms were reported. The Romberg test was measured as time to failure (seconds) in the eyes-closed state and on firm ground to evaluate proprioceptive function [15]. All the parameters were checked by one physiatrist who did not participate in the analysis of MRI and fluoroscopic images for the prevention of bias.

All patients underwent lumbosacral MRI, and the level of central stenotic areas was judged by morphological grading. Stenosis was classified into none, mild, moderate, or severe as described in a previous study [16], and when the spinal canal showed more than moderate stenosis and matched the clinical symptom level, that level of the spine was considered the stenosis level.

2.3 Epidural injection technique

Inter-laminar epidural injection was performed in the lateral decubitus position with the lumbar spine flexed to widen the interspinous area. One percent chlorhexidine and lidocaine solutions were used for sterilization and anesthesia of the skin respectively around the lower lumbar area. After confirmation of the L3-4 level by fluoroscopy (SONIALVISION Safire 2; Shimadzu, Kyoto Japan), a 20-gauge Tuohy needle (Contiplex ${}^{\@setsize{\scriptsize}{8pt}{\viipt}{\@viipt}\textregistered}$ , Braun Medical Inc., PA, USA) was inserted into the interspinous area at L3-4. We set the injection point at the level of L3-4 because midline approach of inter-laminar epidural injection was impossible for several patients due to the kissing spinous processes of L4 and L5. The inter-laminar space of L3-4 has been reported to be the widest in most elderly patients, making the procedure relatively easy [17]. Contrast spread is known to be affected by the injection level [5]. Therefore, we unified the injection point at one fixed interspinous level instead of changing the injection point according to the most stenotic level to find the coverage of stenotic levels and to measure the degree of contrast spread uniformly. When the needle tip reached the epidural space using the loss-of-resistance technique, 1 ml (647 mg of iohexol) of radiopaque contrast (Iobrix ${}^{\@setsize{\scriptsize}{8pt}{\viipt}{\@viipt}\textregistered}$ , Taijoon Corp, Seoul, Korea) was injected slowly under fluoroscopic guidance to confirm epidural spread. A total of 8 ml of 1% lidocaine (Lidocaine HCl ${}^{\@setsize{\scriptsize}{8pt}{\viipt}{\@viipt}\textregistered}$ , Daihan Pharm Corp., Seoul, Korea) mixed with 40 mg of triamcinolone acetate (Tamceton ${}^{\@setsize{\scriptsize}{8pt}{\viipt}{\@viipt}\textregistered}$ , Hanall Biopharm, Seoul, Korea) was then infused slowly. Using the image acquired immediately after the injection of lidocaine and triamcinolone, we measured the degree of contrast spread in the lateral view as a real distance (mm) using PACS image calipers (Fig. 1).

Figure 1.

Inter-laminar epidural injection was performed at the L3-4 level and the degree of contrast spread was measured as a real distance (mm) using PACS image calipers. Contrast was sufficiently distributed over the stenotic area (1A) and insufficiently distributed over the stenotic area (1B).

All patients were asked not to change their current medications to determine the accurate effect of inter-laminar epidural injection. Their medications were nonsteroidal anti-inflammatory drugs or gabapentin or both.

When the contrast covered all of the stenosis levels, this was considered sufficient distribution (SD) (Fig. 1A) and when the contrast did not cover all stenosis levels, this was considered insufficient distribution (ID) (Fig. 1B). To prevent bias, the patients did not receive any information about the groups they belonged to.

2.4 Statistical analysis

To determine the difference in clinical improvement between SD and ID, the difference in ODI values at 2 weeks was set as the primary outcome and subjected to Mann-Whitney U test. Changes in other parameters were also compared between the two groups using Mann-Whitney U test. To determine the correlation between effectiveness and contrast distribution, chi-square tests were performed. To identify the correlation between clinical improvements and contrast distribution, Pearson’s correlation analysis was conducted. The SPSS 22.0 software (IBM SPSS Statistics; IBM Corporation, Somers, NY) was used for the statistical analysis, and a $P$ -value less than 0.05 was considered to indicate statistical significance.

Figure 2.

Stenotic levels that reflected more than moderate stenosis in the MRI were considered stenosis.

3. Results

A total of 24 patients participated in our study, comprising 6 males and 18 females (average height 151.1 cm with one standard deviation 30.0), their average age was 67.5 $\pm$ 12.9 years (mean $\pm$ standard deviation). Time from the initial symptom to the inter-laminar epidural injection was 3.5 $\pm$ 1.1 years. The initial NRS was 6.0 (interquartile range (IQR), 2.0), the initial claudication distance was 200.0 (IQR, 150.0) meter, and the initial ODI was 35.5 (IQR, 17.2).

Central stenotic levels as shown on MRIs ranged from L2-3 to L5-S1; the distribution is presented in Fig. 2. After inter-laminar epidural injection, contrast was distributed along the epidural space at an average distance of 53.2 $\pm$ 17.1 mm (range, 24.7 to 87.0 mm). SD occurred in 13 patients, and ID in 11 patients (Fig. 2). After the injection, there was no patient who complained of complications including vasovagal syncope and headache. None of the patients changed their treatment including oral medication and physical modalities during the study period.

When the patients were divided into two groups (SD and ID), initial NRS, initial claudication distance, initial Romberg test time, and initial ODI were not significantly different between the groups. At 2 weeks, percentages of NRS improvement were 71.4 (IQR, 20.2) in SD group and 50.0 (IQR, 31.4) in ID group ( $P=$ 0.02), changes in Romberg test time were 0.0 (IQR, 35.0) in SD group and $-$ 3.0 (IQR, 12.0) in ID group ( $P=$ 0.02), decreases in ODI were 9.6 (IQR, 11.2) in SD group and 0.0 (IQR, 7.8) in ID group ( $P=$ 0.02), and changes of claudication distance were 200.0 (IQR, 350.0) in SD group and $-$ 100.0 (IQR, 340.0) ( $P>$ 0.05) (Fig. 3). At 6 weeks, percentages of NRS improvement were 50.0 (IQR, 77.5) in SD group and 50.0 (IQR, 80.0) in ID group ( $P>$ 0.05), changes in Romberg test time were $-$ 6.0 (IQR, 15.4) in SD group and 0.0 (IQR, 15.0) in ID group ( $P>$ 0.05), decreases in ODI were 8.9 (IQR, 17.8) in SD group and 0.0 (IQR, 10.0) in ID group ( $P>$ 0.05), and changes of claudication distance were 300.0 (IQR, 230.0) in SD group and 100.0 (IQR, 400.0) ( $P>$ 0.05) (Fig. 4). At 12 weeks, percentages of NRS improvement were 28.6 (IQR, 102.4) in SD group and 37.5 (IQR, 63.4) in ID group ( $P>$ 0.05), changes in Romberg test time were $-$ 5.0 (IQR, 15.0) in SD group and $-$ 2.0 (IQR, 19.0) in ID group ( $P>$ 0.05), decreases in ODI were 0.0 (IQR, 17.0) in SD group and 2.2 (IQR, $-$ 18.2) in ID group ( $P>$ 0.05), and changes of claudication distance were 300.0 (IQR, 240.0) in SD group and 100.0 (IQR, 400.0) ( $P>$ 0.05) (Fig. 5).

Figure 3.

Box plot graph showing the percentage of numeric rating scale (NRS), change of Romberg test, decrease of Oswestry Disability Index (ODI), and change of claudication distance 2 weeks after the inter-laminar epidural injection in sufficient distribution (SD) and insufficient distribution (ID) groups.

Figure 4.

Box plot graph showing the percentage of numeric rating scale (NRS), change of Romberg test, decrease of Oswestry Disability Index (ODI), and change of claudication distance 6 weeks after the inter-laminar epidural injection in sufficient distribution (SD) and insufficient distribution (ID) groups.

Figure 5.

Box plot graph showing the percentage of numeric rating scale (NRS), change of Romberg test, decrease of Oswestry Disability Index (ODI), and change of claudication distance 12 weeks after the inter-laminar epidural injection in sufficient distribution (SD) and insufficient distribution (ID) groups.

Both groups showed significant change of NRS (F ${}_{df=3}=$ 17.55, $P<$ 0.01), significant change of claudication distance (F ${}_{df=3}=$ 3.10, $P=$ 0.03), and significant decrease of ODI (F ${}_{df=3}=$ 3.67, $P=$ 0.02) over time except Romberg test time (F ${}_{df=3}=$ 2.05, $P=$ 0.12).

At 2 weeks after the injection, 7 patients in the SD group showed ‘more effective in NRS’ and 6 patients showed ‘less effective in NRS’. In the ID group, only one patient showed ‘more effective in NRS’ while 10 patients showed ‘less effective in NRS’, which was significantly different ( $\chi^{2}=$ 5.37, $P=$ 0.03) from those in the SD group. However, the results were not significantly different between the two groups at 6 and 12 weeks after the injection. In terms of ODI at 2 weeks, 9 patients in the SD group showed ‘more effective’ and 4 patients showed ‘less effective’, while 2 patients in the ID group showed ‘more effective’ and 9 patients in the ID group showed ‘less effective’, which was significantly different ( $\chi^{2}=$ 6.25, $P=$ 0.02) from those in the SD group. The response in ODI was not significantly different between the two groups at 6 and 12 weeks after the injection (Table 1).

Table 1

Number of patients who showed ‘more effective’ and ‘less effective’ in numeric rating scale (NRS) and Oswestry Disability Index (ODI) at 2 weeks, 6 weeks, and 12 weeks after the interlaminar epidural injection according to the contrast distribution

		SD group	ID group	$P$ -value
NRS at 2 weeks ${}^{*}$	More effective	7	1	0.03
	Less effective	6	10
NRS at 6 weeks	More effective	5	5	1.00
	Less effective	8	6
NRS at 12 weeks	More effective	4	4	1.00
	Less effective	9	7
ODI at 2 weeks ${}^{*}$	More effective	9	2	0.02
	Less effective	4	9
ODI at 6 weeks	More effective	8	3	0.12
	Less effective	5	8
ODI at 12 weeks	More effective	6	4	0.697
	Less effective	7	7

Sufficient distribution (SD) was defined when the contrast covered all of the stenosis levels and insufficient distribution (ID) when the contrast did not cover all stenosis levels. ${}^{*}$ : $P$ -value $<$ 0.05.

The average distance (mm) of the contrast distribution showed no correlation with the percentage of NRS improvement, change of claudication distance, change of Romberg test, or decrease of ODI at any time point; the correlations between the distances of contrast distribution and all clinical parameters at 2 weeks are presented in Fig. 6.

Figure 6.

Correlations between the distance of contrast distribution and changes in the numeric rating scale (NRS), claudication distance, Romberg test, and Oswestry Disability Index (ODI) at 2 weeks. There were no significant correlations between the distance of contrast distribution and any of the parameters.

4. Discussion

In our study, we found that a contrast distribution that covered the stenotic area showed greater short term reduction in pain with the inter-laminar epidural injection than did insufficient distribution over the stenotic area. Moreover, a wide contrast distribution that did not cover the stenotic area did not show better results, emphasizing the importance of the accurate distribution of contrast in spinal stenosis.

Injected materials for inter-laminar epidural injection at L3-4 traveled one or two vertebral levels caudad or cephalad, which always included the most stenotic level in our cases. We used the same volume of injected materials in all patients, but the distribution range varied. This might be due to needle tip position, gender, lumbar lordosis, operative history, and degree of stenotic areas according to previous studies (5, 8, 18). We performed the inter-laminar epidural injection in the lateral position instead of prone position because the fluoroscopy could not be rotated during the procedure.

We used 8 ml of the mixture of lidocaine and triamcinolone acetate in the inter-laminar epidural injection by midline approach. This amount was enough to cover the posterior and anterior parts of epidural space in all the patients, which would reduce the pain from central canal stenosis caused by facet joint hypertrophy or disc problems if the stenotic level was covered by the drugs.

In terms of the correlation of contrast distribution on clinical improvement, our results are consistent with a study by Manchikanti [19] in which inter-laminar epidural injection was found to be superior to caudal epidural injection in treating lumbar central spinal stenosis because caudal epidural injection might not reach the target level. Milburn reported that inter-laminar epidural injection was most effective when performed at the maximal stenotic level, which supports our findings [10]. Cosgrove [7] suggested that successful response was not related to the severity of stenosis on MRI, which contradicts our results. Our study suggested that distribution of contrast to a more than moderate stenotic level was important for pain reduction. The study by Cosgrove [7] and Kuittinen [20] focused on the canal dimension area, not the stenotic level, and this contradiction can be explained by the fact that the symptom was not related to the canal dimension area but the stenotic level.

Four patients in SD group and four patients in ID group had stenotic areas at the level of L3-4, the injection site. In the SD group, stenotic areas ranged from L2-3 to S1-2. In the ID group, the stenotic areas ranged from L3-4 to L5-S1. Therefore, the superior effect in the SD group might not be due to close direction of medications into the stenotic areas.

In our study, all clinical symptoms in both groups were improved over time, with the exception of Romberg test time. A systematic review performed by Ammendolia showed that the evidence for surgical and non-surgical treatment to improve walking ability was of low and very low quality [21], which might explain the inconsistent effect of inter-laminar epidural injection on the claudication distance in our study.

Unexpectedly, Romberg test time in ID group decreased at 2 weeks after the injection, which was significantly different in paired t-test ( $P=$ 0.01). We thought that this paradoxical worsening of the Romberg test was due to the decrease of proprioception by diffuse epidural block of the lidocaine. Akabane et al. [22] proved that epidural block disturbed postural stability, but this disturbance lasted only at a maximum 90 minutes, which does not explain this persistent paradoxical worsening. This paradoxical worsening was severe in three patients who belonged to ID group. We thought that this paradoxical worsening in three patients might be due to the aggravation of spinal canal compromise or other causes, such as, peripheral sensory neuropathy, not the epidural injection.

Median value of claudication distance change in ID group also decreased at 2 weeks after the injection. This paradoxical decrease might be due to the skewness of data and small sample size because its mean value increased to 4.5 meter. Claudication distance between baseline and 2 weeks after the injection was not significantly different in paired t-test ( $P=$ 0.96). Walking speed used for claudication distance was 5 to 6 km per hour, which was faster than that used in the previous study [23]. Rainville et al. [23] reported self-paced walking test demonstrated better internal responsiveness than treadmill test and self-reported measures. We think that the walking speed for claudication distance in our study might affect the results. Later study to use self-paced walking test must be done to clarify the effect of epidural distribution on the claudication distance.

The degree of contrast spread did not affect the clinical improvement in our study. The average distance of contrast spread was 53.2 $\pm$ 4.4 mm in the SD group and 53.2 $\pm$ 5.8 mm in the ID group, and neither was statistically significant. One possible explanation is that as the contrast is distributed more widely, smaller amounts of the steroid and anesthetics reach the pain focus. Indeed, when we restricted the patients to the SD group, the degree of contrast spread showed a negative clinical improvement tendency, although there was no significant correlation. This finding suggests that the degree of contrast spread is not important in terms of pain reduction; coverage of the stenotic area is more important. When the degree of contrast spread according to the response in NRS was analyzed, ‘more effective in NRS’ showed 47.2 mm while ‘less effective in NRS’ showed 56.2 mm. Although the two were not significantly different ( $P=$ 0.23), contrast spread seemed to be inversely related to response in NRS if the coverage of the stenotic area was not considered.

Although all clinical parameters except claudication distance showed greater improvements in the SD group 2 weeks after inter-laminar epidural injection, this superiority was absent at 6 and 12 weeks. Because the role of inter-laminar epidural injection in spinal stenosis is short-term pain relief but its efficacy for long-term pain relief is unclear [24], the absence of a difference in clinical improvement between the SD and ID groups is plausible.

4.1 Limitations

Our study had several limitations. First, because significant dilution of the contrast (1:10) was used for contrast dispersion, some medications might have reached the stenotic level and been undetected in the ID group, resulting in NRS improvement, which might have decreased the possibility of detecting the difference.

5. Conclusion

We demonstrated that coverage of the stenotic area by contrast in inter-laminar epidural injection was important for acute pain reduction. Our results also revealed that contrast spread had a negative tendency of ‘more effective’ in pain reduction if the coverage of the stenotic area was not considered. To increase the success rate of inter-laminar epidural injection in lumbar spinal stenosis, the drugs must cover more than a moderate degree of the stenotic area, which can be achieved through optimal needle position or use of an epidural catheter.

Footnotes

Acknowledgments

This work was supported by the grant of the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (grant number: HI16C3229) and ICT R&D program of MSIP/IIPT [Grant No. B0101-16-1081].

Conflict of interest

None to report.

References

Siebert

Pruss

Klingebiel

Failli

Einhaupl

Schwab

. Lumbar spinal stenosis: syndrome, diagnostics and treatment. Nat Rev Neurol. 2009; 5(7): 392-403.

Lee

Kim

Lee

Park

. Lumbar Stenosis: A Recent Update by Review of Literature. Asian Spine J. 2015; 9(5): 818-28.

Ammendolia

Stuber

de Bruin

Furlan

Kennedy

Rampersaud

, et al. Nonoperative treatment of lumbar spinal stenosis with neurogenic claudication: a systematic review. Spine (Phila Pa1976). 2012; 37(10): E609-16.

Parr

Diwan

Abdi

. Lumbar interlaminar epidural injections in managing chronic low back and lower extremity pain: a systematic review. Pain Physician. 2009; 12(1): 163-88.

Weil

Frauwirth

Amirdelfan

Grant

Rosenberg

. Fluoroscopic analysis of lumbar epidural contrast spread after lumbar interlaminar injection. Arch Phys Med Rehabil. 2008; 89(3): 413-6.

Landa

Kim

. Outcomes of interlaminar and transforminal spinal injections. Bull NYU Hosp Jt Dis. 2012; 70(1): 6-10.

Cosgrove

Bertolet

Chase

Cosgrove

. Epidural steroid injections in the treatment of lumbar spinal stenosis efficacy and predictability of successful response. Am J Phys Med Rehabil. 2011; 90(12): 1050-5.

Cleary

Keating

Poynton

. The flow patterns of caudal epidural in upper lumbar spinal pathology. Eur Spine J. 2011; 20(5): 804-7.

Campbell

Carreon

Glassman

McGinnis

Elmlinger

. Correlation of spinal canal dimensions to efficacy of epidural steroid injection in spinal stenosis. J Spinal Disord Tech. 2007; 20(2): 168-71.

10.

Milburn

Freeman

Steven

Altmeyer

Kay

. Interlaminar epidural steroid injection for degenerative lumbar spinal canal stenosis: does the intervertebral level of performance matter? Ochsner J. 2014; 14(1): 62-6.

11.

Ghai

Vadaje

Wig

Dhillon

. Lateral parasagittal versus midline interlaminar lumbar epidural steroid injection for management of low back pain with lumbosacral radicular pain: a double-blind, randomized study. Anesth Analg. 2013; 117(1): 219-27.

12.

Bosscher

Heavner

. Incidence and severity of epidural fibrosis after back surgery: an endoscopic study. Pain Pract. 2010; 10(1): 18-24.

13.

Fairbank

Pynsent

. The Oswestry Disability Index. Spine (Phila Pa 1976). 2000; 25(22): 2940-52; discussion 52.

14.

Frost

Lamb

Stewart-Brown

. Responsiveness of a patient specific outcome measure compared with the Oswestry Disability Index v21 and Roland and Morris Disability Questionnaire for patients with subacute and chronic low back pain. Spine (Phila Pa 1976). 2008; 33(22): 2450-7; discussion 8.

15.

Agrawal

Carey

Hoffman

Sklare

Schubert

. The modified Romberg Balance Test: normative data in U.S. adults. Otol Neurotol. 2011; 32(8): 1309-11.

16.

Lee

Choi

Kang

. A new grading system of lumbar central canal stenosis on MRI: an easy and reliable method. Skeletal Radiol. 2011; 40(8): 1033-9.

17.

Nagaro

Yorozuya

Kamei

Kii

Arai

Abe

. Fluoroscopically guided epidural block in the thoracic and lumbar regions. Reg Anesth Pain Med. 2006; 31(5): 409-16.

18.

Whitlock

Bridwell

Gilula

. Influence of needle tip position on injectate spread in 406 interlaminar lumbar epidural steroid injections. Radiology. 2007; 243(3): 804-11.

19.

Manchikanti

Falco

Pampati

Hirsch

. Lumbar interlaminar epidural injections are superior to caudal epidural injections in managing lumbar central spinal stenosis. Pain Physician. 2014; 17(6): E691-702.

20.

Kuittinen

Sipola

Saari

Aalto

Sinikallio

Savolainen

, et al. Visually assessed severity of lumbar spinal canal stenosis is paradoxically associated with leg pain and objective walking ability. BMC Musculoskelet Disord. 2014; 15: 348.

21.

Ammendolia

Stuber

Tomkins-Lane

Schneider

Rampersaud

Furlan

, et al. What interventions improve walking ability in neurogenic claudication with lumbar spinal stenosis? A systematic review. Eur Spine J. 2014; 23(6): 1282-301.

22.

Akabane

Shimada

Ogawa

. Usefulness of posturography after epidural block. J Nippon Med Sch. 2004; 71(1): 35-43.

23.

Rainville

Childs

Pena

Suri

Limke

Jouve

, et al. Quantification of walking ability in subjects with neurogenic claudication from lumbar spinal stenosis – a comparative study. Spine J. 2012; 12(2): 101-9.

24.

Manchikanti

Kaye

Manchikanti

Boswell

Pampati

Hirsch

. Efficacy of epidural injections in the treatment of lumbar central spinal stenosis: a systematic review. Anesth Pain Med. 2015; 5(1): e23139.