new england

The
journal of medicine
established in 1812 july 3, 2014 vol. 371 no. 1

A Randomized Trial of Epidural Glucocorticoid Injections

for Spinal Stenosis
Janna L. Friedly, M.D., Bryan A. Comstock, M.S., Judith A. Turner, Ph.D., Patrick J. Heagerty, Ph.D.,
Richard A. Deyo, M.D., M.P.H., Sean D. Sullivan, Ph.D., Zoya Bauer, M.D., Ph.D., Brian W. Bresnahan, Ph.D.,
Andrew L. Avins, M.D., M.P.H., Srdjan S. Nedeljkovic, M.D., David R. Nerenz, Ph.D., Christopher Standaert, M.D.,
Larry Kessler, Ph.D., Venu Akuthota, M.D., Thiru Annaswamy, M.D., Allen Chen, M.D., M.P.H., Felix Diehn, M.D.,
William Firtch, M.D., Frederic J. Gerges, M.D., Christopher Gilligan, M.D., Harley Goldberg, M.D.,
David J. Kennedy, M.D., Shlomo Mandel, M.D., Mark Tyburski, M.D., William Sanders, M.D., David Sibell, M.D.,
Matthew Smuck, M.D., Ajay Wasan, M.D., Lawrence Won, M.D., and Jeffrey G. Jarvik, M.D., M.P.H.

A BS T R AC T

Background
Epidural glucocorticoid injections are widely used to treat symptoms of lumbar The authors affiliations are listed in the
Appendix. Address reprint requests to
spinal stenosis, a common cause of pain and disability in older adults. However, Dr.Friedly at the Department of Rehabili-
rigorous data are lacking regarding the effectiveness and safety of these injections. tation Medicine, University of Washington,
325 Ninth Ave., Seattle, WA 98104, or at
[email protected].
Methods
In a double-blind, multisite trial, we randomly assigned 400 patients who had This article was updated on July 3, 2014, at
NEJM.org.
lumbar central spinal stenosis and moderate-to-severe leg pain and disability to
N Engl J Med 2014;371:11-21.
receive epidural injections of glucocorticoids plus lidocaine or lidocaine alone.
DOI: 10.1056/NEJMoa1313265
The patients received one or two injections before the primary outcome evalua- Copyright 2014 Massachusetts Medical Society.
tion, performed 6weeks after randomization and the first injection. The primary
outcomes were the score on the RolandMorris Disability Questionnaire (RMDQ,
in which scores range from 0 to 24, with higher scores indicating greater physical
disability) and the rating of the intensity of leg pain (on a scale from 0 to 10, with
0 indicating no pain and 10indicating pain as bad as you can imagine).

Results
At 6 weeks, there were no significant between-group differences in the RMDQ score
(adjusted difference in the average treatment effect between the glucocorticoid
lidocaine group and the lidocaine-alone group, 1.0 points; 95% confidence in-
terval [CI], 2.1 to 0.1; P=0.07) or the intensity of leg pain (adjusted difference in
the average treatment effect, 0.2 points; 95% CI, 0.8 to 0.4; P=0.48). A pre-
specified secondary subgroup analysis with stratification according to type of
injection (interlaminar vs. transforaminal) likewise showed no significant differ-
ences at 6 weeks.

Conclusions
In the treatment of lumbar spinal stenosis, epidural injection of glucocorticoids
plus lidocaine offered minimal or no short-term benefit as compared with epi-
dural injection of lidocaine alone. (Funded by the Agency for Healthcare Research
and Quality; ClinicalTrials.gov number, NCT01238536.)

n engl j med 371;1nejm.orgjuly 3, 2014 11

The New England Journal of Medicine
Downloaded from nejm.org on December 3, 2014. For personal use only. No other uses without permission.
Copyright 2014 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

L
umbar spinal stenosis, a common Enrollment and Randomization
cause of spine-related disability, is the lead- Patients with spinal stenosis who were referred
ing reason for spinal surgery in older for epidural glucocorticoid injections were invit-
adults.1,2 Degenerative changes resulting in nar- ed to participate in this study by the treating phy-
rowing of the spinal canal and nerve-root com- sician or research coordinator. To be eligible,
pression can cause back and leg pain, lower- participants had to be at least 50 years of age and
extremity paresthesias, and weakness.3,4 The have evidence of central lumbar spinal stenosis
treatment of symptomatic lumbar stenosis re- on magnetic resonance imaging or computed to-
mains controversial. mography. Additional eligibility criteria were an
Symptoms of lumbar stenosis are commonly average pain rating of more than 4 (on a scale of
treated with epidural glucocorticoid injections. 0 to 10, with 0 indicating no pain and 10 indicat-
These injections typically contain a glucocorti- ing pain as bad as you can imagine) for pain in
coid and an anesthetic, which are thought to the lower back, buttock, leg, or a combination of
relieve pain by reducing nerve-root inflamma- these sites on standing, walking, or spinal exten-
tion and ischemia.1 An estimated 25% of all sion in the past week; worse pain in the buttock,
epidural glucocorticoid injections administered leg, or both than in the back; and a score of 7 or
in the Medicare population and 74% of those higher on the RolandMorris Disability Question-
administered in patients at Veterans Affairs naire (RMDQ) (with scores ranging from 0 to 24,
medical centers are for spinal stenosis.5,6 Rates and higher scores indicating greater disability).
and associated costs of these injections for spi- Patients who did not have stenosis of the central
nal stenosis increased nearly 300% over the past canal, those with spondylolisthesis requiring
two decades, and it is estimated that more than surgery, and those who had a history of lumbar
2.2 million lumbar epidural glucocorticoid injec- surgery or had received epidural glucocorticoid
tions are performed each year in the Medicare injections within the previous 6 months were ex-
population.5-8 Uncontrolled studies suggest that cluded. Additional inclusion and exclusion crite-
these injections provide short-term pain relief for ria are provided in Table 1 in the Supplementary
at least some patients with spinal stenosis,9-15 Appendix, available at NEJM.org.
but data are lacking from rigorous randomized, A study physician reviewed each potential
controlled trials evaluating effectiveness and participants imaging studies to confirm the pres-
safety.16 Therefore, we designed the Lumbar ence of stenosis in the central spinal canal and
Epidural Steroid Injections for Spinal Stenosis rated it as mild, moderate, or severe. Patients were
(LESS) trial to compare the effectiveness of epi- informed that they would be randomly assigned
dural injections of glucocorticoids plus anesthetic to receive a standard epidural injection of either
with injections of anesthetic alone in patients glucocorticoids plus lidocaine or lidocaine alone,
with lumbar spinal stenosis. that they could receive a repeat injection at 3 weeks
if they wished (at the discretion of the treating
Me thods physician), and that they could cross over to the
other treatment after the 6-week assessment.
Study Design and Oversight The data coordinating center generated elec-
This double-blind, randomized, controlled trial tronically concealed, permuted-block randomized
was conducted at 16 sites in the United States. assignments for each recruitment site. Two
Details of the trial design and methods were opaque syringes one with glucocorticoids
published previously.17 The study was approved plus lidocaine and the other with lidocaine alone
by the institutional review board at each site and were prefilled for each procedure. At the time
was overseen by an independent data and safety of the procedure, the randomization assignment
monitoring board. All patients provided written was obtained through a password-protected study
informed consent. There was no commercial website by a clinical assistant who was not in-
sponsorship. The first and last authors vouch for volved with subsequent data collection. The
the accuracy of the data and analyses and adher- assignment indicated which syringe should be
ence to the study protocol (available with the full labeled inject and which should be labeled
text of this article at NEJM.org). This report is discard. The assistant confirmed the physicians
focused on the patient-reported outcomes at the use of the syringe marked inject. In this blinded
primary evaluation 6 weeks after randomization. trial, the treating physicians, patients, and re-

12 n engl j med 371;1nejm.orgjuly 3, 2014

The New England Journal of Medicine

Downloaded from nejm.org on December 3, 2014. For personal use only. No other uses without permission.
Copyright 2014 Massachusetts Medical Society. All rights reserved.
 Glucocorticoid Injections for Spinal Stenosis

search staff who conducted follow-up were un- from 0 to 24, with higher scores indicating more
aware of the treatment received. depressive symptoms)23; scores on the Generalized
Anxiety Disorder 7 scale (GAD-7; on a scale from
Interventions 0 to 21, with higher scores indicating more severe
Twenty-six board-certified anesthesiologists, phys- anxiety)24; scores on the EQ-5D (formerly called the
iatrists, and radiologists with expertise in admin- European Quality of Life5 Dimensions) question-
istering epidural glucocorticoid injections per- naire (with scores ranging from 0 to 1 and lower
formed the procedures. Study physicians were scores indicating worse quality of life)25; and
trained by the study investigators to administer scores on the Swiss Spinal Stenosis Questionnaire
the injections in a standardized manner with the (SSSQ),26 which was composed of subscales for
use of fluoroscopic guidance.17 Physicians were symptoms (scores of 1 to 5, with higher scores
instructed to choose the injection level (e.g., L5S1) indicating worse symptoms), physical function
one spinal level below the maximal canal steno- (scores of 1 to 4, with higher scores indicating
sis for interlaminar injections and at the root worse function), and satisfaction with treatment
level where symptoms were most pronounced for (scores of 1 to 4, with higher scores indicating
transforaminal injections, although bilateral and less satisfaction).
multilevel transforaminal injections were allowed.
The physician chose the approach (transforami- Assessments
nal or interlaminar), which remained consistent At baseline, patients provided demographic, clin-
with subsequent injections for each patient. The ical, and disease-history information and rated
glucocorticoid injectable solution consisted of 1 to their expectations of pain relief after receiving an
3 ml of 0.25% to 1% lidocaine followed by 1 to 3 ml injection. All outcomes were assessed at baseline
of triamcinolone (60 to 120 mg), betamethasone and at 3 and 6 weeks after randomization. Re-
(6 to 12 mg), dexamethasone (8 to 10 mg), or search staff who were unaware of the patients
methylprednisolone (60 to 120 mg). The physi- treatment assignments conducted all assessments
cian selected the glucocorticoid according to his by telephone, in-person interview, or mailed
or her usual practice. The procedure for the lido- questionnaires. Five weeks after randomization,
caine injection was identical to that for the glu- the patients and providers were asked to guess
cocorticoidlidocaine injection except that the which treatment the patient had received. Morn-
injectable solution was an equivalent volume of ing serum cortisol levels (before 9 a.m.) were
0.25% to 1% lidocaine alone. also obtained at baseline, 3 weeks, and 6 weeks
as a measure of the degree of systemic glucocor-
Outcomes ticoid absorption.
Two primary outcomes were measured at 6 weeks:
the RMDQ score18 and the patients rating of av- Adverse-Event Reporting
erage buttock, hip, or leg pain in the previous Participants were asked at each follow-up about un-
week. The RMDQ wording was modified slightly expected medical events, including hospitalizations
to specify problems due to back pain or leg pain and emergency department visits, since the last
(sciatica) rather than solely back pain. contact. Treating physicians documented adverse
Secondary outcomes included the proportion events during or immediately after the procedure.
of patients with at least minimal clinically mean-
ingful improvement (30%) and the proportion Statistical Analysis
with substantial clinically meaningful improve- For the primary analyses, we used an intention-
ment (50%)19-21 from baseline to 6 weeks on to-treat strategy. The differences in treatment
each measure. Other secondary outcomes were effects and 95% confidence intervals were calcu-
ratings of average back pain in the previous lated with the use of analysis of covariance
week (on a scale of 0 to 10, with higher scores (ANCOVA) with adjustment for the baseline value
indicating a higher intensity of back pain); scores of the outcome measure. Adjustment was also
on the Brief Pain Inventory (BPI) interference made for the recruitment site. An indicator of
scale (from 0 to 10, with higher scores indicat- treatment group was coded such that negative
ing more pain-related interference with activi- values of the treatment effect indicated greater
ty)22; scores on the eight-question version of the improvement in the glucocorticoidlidocaine
Patient Health Questionnaire (PHQ-8; on a scale group. We conducted parallel analyses of RMDQ

n engl j med 371;1nejm.orgjuly 3, 2014 13

The New England Journal of Medicine
Downloaded from nejm.org on December 3, 2014. For personal use only. No other uses without permission.
Copyright 2014 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

sion models to compare the proportions of pa-

tients in each group who had 30% or more im-
2224 Patients were assessed for eligibility
provement and 50% or more improvement in
RMDQ scores and ratings of the intensity of leg
1824 Did not undergo random- pain, had scores on the SSSQ satisfaction scale
ization indicating satisfaction with treatment, had one
348 Declined screening
1435 Were ineligible or more adverse events in the first 6 weeks, and
437 Had prior back surgery had morning cortisol levels that were consistent
422 Received epidural gluco-
corticoids in past 6 mo with suppression of the hypothalamicpituitary
105 Had less leg pain than axis (Table 3 in the Supplementary Appendix).
back pain
48 Had coexisting We used Poisson regression models with robust
conditions standard errors to compare the 6-week rates of
48 Did not speak English
375 Had other exclusion adverse events per person. The blinding index
criteria described by Bang et al.,27 which ranges from 1
22 Were eligible but declined
to participate to 1, with 1 indicating that all patients guessed
19 Withdrew from study before the incorrect treatment, 0 indicating that all pa-
randomization
tients randomly guessed their study-group assign-
ments, and 1 indicating that all patients correctly
400 Underwent randomization guessed their study-group assignments, was used
to evaluate the effectiveness of blinding within
each treatment group.
We used a Bonferroni-corrected significance
threshold of 0.05 divided by 2 (or 0.025) in
200 Were assigned to and received 200 Were assigned to and received
lidocaine alone glucocorticoids plus lidocaine analyses within subgroups for race and injection
139 Had interlaminar approach 143 Had interlaminar approach approach. All other reported P values are based
61 Had transforaminal approach 57 Had transforaminal approach
on two-sided tests, with a P value of less than
0.05 considered to indicate statistical significance.
189 (94.5%) Completed 3-wk assess- 195 (97.5%) Completed 3-wk assess-
We powered the study to evaluate the overall
ment ment effectiveness of glucocorticoids plus lidocaine
9 Were lost to follow-up 3 Were lost to follow-up versus lidocaine alone, as well as the effective-
2 Withdrew from study 2 Withdrew from study
75 (37.5%) Received an additional 80 (40.0%) Received an additional ness within subgroups based on race (white or
study injection (lidocaine) study injection (glucocorticoid) nonwhite) and injection approach (interlaminar
1 (0.5%) Received glucocorticoids 0 Received lidocaine
or transforaminal). The study was powered for
the outcome of the RMDQ score at 6 weeks,
193 (96.5%) Completed 6-wk assess- 193 (96.5%) Completed 6-wk assess- conservatively assuming a standard deviation of
ment ment 7.5 in each group, a correlation between baseline
5 Were lost to follow-up 5 Were lost to follow-up
2 Withdrew from study 2 Withdrew from study and 6-week scores of 0.30, a retention rate of 85%,
and a two-sided alpha level of 0.05. Assuming a
Figure 1. Enrollment, Randomization, and Follow-up. common treatment effect among the subgroups
in an ANCOVA model, with 400 randomly as-
signed participants, the overall comparison in
an assessment of glucocorticoids plus lidocaine
scores and leg pain and did not prespecify a correc- versus lidocaine alone would have 83% power to
tion for multiple comparisons. Separate ANCOVA detect a difference of 2.25 points or more in the
models were used to estimate the differences RMDQ score (a conservative estimate of the mini-
within subgroups defined according to race mal clinically important difference for the RMDQ
(white or nonwhite) and injection approach (inter- score28) and a difference of 3.25 points or more
laminar or transforaminal). Because there was for prespecified subgroups. With the use of the
an imbalance between groups at baseline with observed study data in a post hoc power analy-
respect to the duration of pain, we conducted sis, the study design had 80% power to detect
apost hoc analysis of the primary outcomes at differences between the groups of 1.6 points in
6 weeks, adjusting for baseline duration of pain. the RMDQ score and 0.8 points in the rating of
In separate analyses, we used logistic-regres- leg pain.

14 n engl j med 371;1nejm.orgjuly 3, 2014

The New England Journal of Medicine

Downloaded from nejm.org on December 3, 2014. For personal use only. No other uses without permission.
Copyright 2014 Massachusetts Medical Society. All rights reserved.
 Glucocorticoid Injections for Spinal Stenosis

R e sult s (200 patients) or lidocaine alone (200 patients).

A total of 282 patients received interlaminar injec-
Patients tions and 118 received transforaminal injections
Between April 2011 and June 2013, a total of 2224 (Fig. 1). The two groups were similar with respect to
patients were screened, of whom 441 were eligible baseline characteristics, except that the duration of
and 400 were randomly assigned to receive an epi- pain was shorter in the lidocaine-alone group (Ta-
dural injection of glucocorticoids plus lidocaine ble 1, and Table 2 in the Supplementary Appendix).

Table 1. Baseline Characteristics of the Study Participants.*

Lidocaine GlucocorticoidLidocaine
Characteristic (N=200) (N=200)
Age yr 68.110.2 68.09.8
Female sex no. (%) 104 (52.0) 117 (58.5)
Race or ethnic group no. (%)
White 139 (69.5) 137 (68.5)
Black 52 (26.0) 53 (26.5)
Other 9 (4.5) 10 (5.0)
Hispanic 6 (3.0) 11 (5.5)
Educational level no. (%)
High school, GED diploma, or less 67 (33.5) 60 (30.0)
Some college, vocational, or technical education 56 (28.0) 72 (36.0)
Undergraduate degree 32 (16.0) 33 (16.5)
Professional or graduate degree 45 (22.5) 35 (17.5)
Married or living with partner no. (%) 111 (55.5) 126 (63.0)
Employment status no. (%)
Employed full-time or part-time 71 (35.5) 57 (28.5)
Retired, not disabled 88 (44.0) 93 (46.5)
Retired, disabled 23 (11.5) 31 (15.5)
Other 18 (9.0) 19 (9.5)
Current smoker no. (%) 32 (16.1) 25 (12.5)
Diabetes, receiving insulin no. (%) 15 (7.5) 16 (8.0)
Body-mass index 29.76.0 31.16.5
Duration of pain no. (%)
<3 mo 40 (20.1) 24 (12.0)
3 to <12 mo 62 (31.2) 59 (29.5)
1 to 5 yr 42 (21.1) 67 (33.5)
>5 yr 55 (27.6) 50 (25.0)
Rating for expectation of pain relief 7.81.9 7.71.8
RMDQ score** 15.74.3 16.14.5
Score on numerical rating scale for intensity of leg pain 7.21.8 7.21.9

* Plusminus values are means SD. GED denotes General Educational Development.
Race or ethnic group was self-reported. The categories of race and ethnic group are not mutually exclusive.
Data were missing for one patient in the lidocaine group.
The body-mass index is the weight in kilograms divided by the square of the height in meters.
The difference between the groups was significant (P=0.02 by the chi-square test).
Patients rated their expectation of pain relief from the epidural injection on a scale from 0 to 10, with 0 indicating not
at all helpful and 10 indicating extremely helpful.
** The composite score on the RolandMorris Disability Questionnaire (RMDQ) is the numerical sum of 24 yes-or-no
items. Total scores range from 0 to 24, with 0 indicating no physical limitations and 24 indicating extreme physical
limitations from back pain, leg pain, or both.
Scores on the numerical rating scale for the intensity of buttock, hip, and leg pain range from 0 to 10, with 0 indicating
no pain and 10 indicating pain as bad as you can imagine.

n engl j med 371;1nejm.orgjuly 3, 2014 15

The New England Journal of Medicine
Downloaded from nejm.org on December 3, 2014. For personal use only. No other uses without permission.
Copyright 2014 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

Outcomes tween-group difference in the RMDQ score (ad-

At 6 weeks, both the glucocorticoidlidocaine justed difference in the average treatment effect
group and the lidocaine-alone group had im- between the glucocorticoidlidocaine group and
provement in the RMDQ score as compared with the lidocaine-alone group, 1.0 points; 95% con-
the baseline score (4.2 points and 3.1 points, fidence interval [CI], 2.1 to 0.1; P=0.07) or the
respectively), but there was no significant be- intensity of leg pain (adjusted difference in the av-

Table 2. Primary Outcomes According to Treatment Group and Injection Approach.*

Variable Lidocaine GlucocorticoidLidocaine Treatment Comparison

Mean Mean
Change Change Adjusted
No. of Overall from No. of Overall from Difference
Patients Mean Baseline Patients Mean Baseline (95% CI) P Value
Overall
RMDQ score
Baseline 200 15.74.3 200 16.14.5
3 wk 189 13.15.7 2.64.4 195 11.76.1 4.45.7 1.8 (2.8 to 0.9) <0.001
6 wk 193 12.56.4 3.15.3 193 11.86.3 4.25.8 1.0 (2.1 to 0.1) 0.07
Score on numerical rating scale for
leg pain
Baseline 200 7.21.8 200 7.21.9
3 wk 188 5.02.8 2.22.9 195 4.42.7 2.92.8 0.6 (1.2 to 0.1) 0.02
6 wk 193 4.62.9 2.63.0 193 4.42.9 2.83.1 0.2 (0.8 to 0.4) 0.48
Interlaminar approach
RMDQ score
Baseline 139 16.04.1 143 16.74.3
3 wk 135 13.25.6 2.94.6 139 11.36.3 5.45.8 2.5 (3.7 to 1.3) <0.001
6 wk 136 12.66.3 3.35.3 136 11.86.5 4.86.0 1.4 (2.8 to 0.1) 0.04
Score on numerical rating scale for leg
pain
Baseline 139 7.41.8 143 7.31.9
3 wk 134 5.02.7 2.43.0 139 4.12.7 3.22.9 0.9 (1.5 to 0.3) 0.005
6 wk 136 4.52.9 2.83.1 136 4.23.0 3.13.3 0.3 (1.0 to 0.4) 0.37
Transforaminal approach
RMDQ score
Baseline 61 14.84.5 57 14.44.4
3 wk 54 13.06.1 1.83.9 56 12.65.4 1.84.7 0.1 (1.7 to 1.6) 0.94
6 wk 57 12.16.6 2.65.3 57 12.05.6 2.44.7 0.3 (1.9 to 1.8) 0.95
Score on numerical rating scale for leg
pain
Baseline 61 7.01.8 57 7.02.0
3 wk 54 5.12.7 2.02.6 56 5.02.5 1.92.1 0.0 (0.9 to 0.9) 0.99
6 wk 57 4.92.7 2.02.8 57 4.92.6 2.02.6 0.1 (0.9 to 1.0) 0.89

* Plusminus values are means SD. CI denotes confidence interval.

Differences were adjusted for baseline outcome values as well as recruitment site. Negative analysis-of-covariance coefficients favor the gluco
corticoidlidocaine group; positive coefficients favor the lidocaine-only group.
In the primary outcome assessment, P<0.05 was considered to indicate statistical significance.
In the prespecified subgroup analysis, P<0.025 was considered to indicate statistical significance.

16 n engl j med 371;1nejm.orgjuly 3, 2014

The New England Journal of Medicine

Downloaded from nejm.org on December 3, 2014. For personal use only. No other uses without permission.
Copyright 2014 Massachusetts Medical Society. All rights reserved.
 Glucocorticoid Injections for Spinal Stenosis

erage treatment effect, 0.2 points; 95% CI, 0.8 at 3 or 6 weeks (Table 2). There were no signifi-
to 0.4; P=0.48) (Table 2). At 3 weeks, there were cant interactions between race and treatment in
small between-group differences in the RMDQ analyses of RMDQ scores (P=0.73 for interaction)
score (average treatment effect, 1.8 points; or leg pain (P=0.99 for interaction) at 6 weeks.
95%CI, 2.8 to 0.9; P<0.001) and the intensity of
leg pain (average treatment effect, 0.6 points; Adverse Events
95% CI, 1.2 to 0.1; P=0.02). There were no sig- The proportion of patients reporting one or more
nificant differences in the proportions of patients adverse events was 21.5% in the glucocorticoid
in the glucocorticoidlidocaine group and the lidocaine group and 15.5% in the lidocaine-alone
lidocaine-alone group who had 30% improve- group (P=0.08). There were more adverse events
ment in the RMDQ score (37.3% and 31.6%, re- on average per person in the glucocorticoidlido-
spectively; P=0.24), 50% improvement in the caine group than in the lidocaine-alone group
RMDQ score (23.8% and 20.2%, P=0.39), 30% im- (P=0.02) (Table 3). Among the patients who re-
provement in the rating of leg pain at 6 weeks ceived glucocorticoids plus lidocaine, the rate of
(49.2% and 49.7%, P=0.88), and 50% improvement adverse events was higher among patients who
in the rating of leg pain at 6 weeks (38.3% and received transforaminal injections (0.46) than
38.3%, P=0.97). Post hoc adjustment for the base- among patients who received interlaminar injec-
line duration of pain resulted in a statistically sig- tions (0.22) (Table 3, and Table 3 in the Supple-
nificant but small between-group difference in the mentary Appendix).
RMDQ score (average treatment effect, 1.2 points;
95% CI, 2.3 to 0.1; P=0.03) at 6 weeks, but Cortisol Suppression
with no significant between-group difference in At both 3 and 6 weeks, a significantly higher
the intensity of leg pain (average treatment ef- proportion of patients in the glucocorticoid
fect, 0.3 points; 95% CI, 0.9 to 0.3; P=0.32). lidocaine group than in the lidocaine-alone
group had morning serum cortisol levels of less
Secondary Outcomes than 3g per deciliter (<80 nmol per liter) or
At 6 weeks, there were no significant differences less than 10 g per deciliter (300 nmol per liter)
between the treatment groups with respect to the (Table 3).
BPI, SSSQ symptoms and physical function, EQ-5D,
or GAD-7 scales (Fig. 2). On the PHQ-8 scale, the Blinding
glucocorticoidlidocaine group had more improve- The physicians, patients, and outcome assessors
ment with respect to symptoms of depression were unaware of the assigned treatment (blind-
(P=0.007). On the SSSQ satisfaction scale, 67% of ing index in the glucocorticoidlidocaine group,
patients who received glucocorticoids plus lido- 0.04; 95% CI, 0.02 to 0.09; blinding index in
caine reported being very or somewhat satisfied the lidocaine-alone group, 0.04; 95% CI, 0.02
with their treatment, as compared with 54% of to 0.10).
those who received lidocaine alone (P=0.01).
Discussion
Subgroup Analyses
Among patients who received interlaminar injec- In a multicenter, double-blind trial of fluoro-
tions, those assigned to glucocorticoids plus li- scopically guided epidural injections for lumbar
docaine, as compared with those assigned to li- spinal stenosis, we observed no significant dif-
docaine alone, reported better physical function ferences at 6 weeks between patients assigned to
on the RMDQ (average treatment effect, 2.5 glucocorticoids plus lidocaine and those as-
points; 95% CI, 3.7 to 1.3; P<0.001) and less signed to lidocaine alone with respect to pain-
leg pain (average treatment effect, 0.9 points; related functional disability (as measured by the
95% CI, 1.5 to 0.3; P=0.005) at 3 weeks. How- RMDQ) or pain intensity. At 3 weeks, the gluco-
ever, there were no significant differences be- corticoidlidocaine group had greater improve-
tween the two treatment groups in any outcome ment than the lidocaine-alone group, but the dif-
at 6 weeks. Among patients who received trans- ferences were clinically insignificant.
foraminal injections, there were no significant Despite a rapid increase in the use of epi-
differences between the groups in any outcome dural glucocorticoid injections for lumbar spinal

n engl j med 371;1nejm.orgjuly 3, 2014 17

The New England Journal of Medicine
Downloaded from nejm.org on December 3, 2014. For personal use only. No other uses without permission.
Copyright 2014 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

A Intensity of Back Pain B BPI Interference Scale

10 Baseline 3 Wk 6 Wk 10 Baseline 3 Wk 6 Wk
Lidocaine 6.62.6 4.62.6 4.42.7 Lidocaine 5.63.0 3.92.7 3.83.1
8 Glucocorticoid 6.72.3 4.02.7 4.32.8 8 Glucocorticoid 6.02.3 3.62.9 3.52.9
Mean Score

Mean Score
ATE at 6 wk, 0.1 (95% CI, 0.7 to 0.4) ATE at 6 wk, 0.4 (95% CI, 1.0 to 0.2)
6 6
P=0.58 P=0.17
4 4

2 2

0 0
Baseline 3 Wk 6 Wk Baseline 3 Wk 6 Wk

C SSSQ Symptoms Subscale D SSSQ Physical-Function Subscale

5 Baseline 3 Wk 6 Wk 4.0 Baseline 3 Wk 6 Wk
Lidocaine 3.10.6 2.70.7 2.60.8 Lidocaine 2.50.5 2.20.6 2.20.6
Glucocorticoid 3.20.6 2.50.7 2.50.7 3.5 Glucocorticoid 2.50.5 2.20.6 2.30.7
4
Mean Score

Mean Score
ATE at 6 wk, 0.1 (95% CI, 0.2 to 0.0) 3.0 ATE at 6 wk, 0.1 (95% CI, 0.1 to 0.2)
P=0.18 P=0.40
3 2.5
2.0
2
1.5
1 1.0
Baseline 3 Wk 6 Wk Baseline 3 Wk 6 Wk

E EQ5D F PHQ-8
25 Baseline 3 Wk 6 Wk
1.0
Lidocaine 6.15.5 4.54.6 4.85.1
0.8 20 Glucocorticoid 7.15.7 5.15.4 4.44.3
Mean Score

Mean Score

15 ATE at 6 wk, 1.0 (95% CI, 1.7 to 0.3)

0.6 P=0.007
Baseline 3 Wk 6 Wk
0.4 Lidocaine 0.590.20 0.680.19 0.680.19 10
Glucocorticoid 0.570.20 0.720.18 0.700.20
0.2 ATE at 6 wk, 0.03 (95% CI, 0.01 to 0.07) 5
P=0.11
0.0 0
Baseline 3 Wk 6 Wk Baseline 3 Wk 6 Wk

G GAD-7
20 Baseline 3 Wk 6 Wk
Lidocaine 4.75.4 3.24.5 3.54.5
15 Glucocorticoid 4.74.7 3.44.6 3.24.2
Mean Score

ATE at 6 wk, 0.3 (95% CI, 1.0 to 0.4)

10 P=0.44

0
Baseline 3 Wk 6 Wk

Figure 2. Secondary Outcomes.

Secondary outcomes were the mean (SD) scores on the numerical rating scale for the intensity of back pain (Panel A), the Brief Pain Inventory
(BPI) interference scale (Panel B), the Swiss Spinal Stenosis Questionnaire (SSSQ) subscales for symptoms (Panel C) and physical function
(Panel D), the EQ-5D (formerly called the European Quality of Life5 Dimensions) questionnaire (Panel E), the eight-question version of the Pa-
tient Health Questionnaire (PHQ-8) (Panel F), and the Generalized Anxiety Disorder 7 (GAD-7) scale (Panel G). The numerical rating scale for
the intensity of back pain indicates ratings of average back pain in the past week (on a scale of 0 to 10, with higher scores indicating a higher in-
tensity of back pain). Scores on the BPI interference scale range from 0 to 10, with higher scores indicating greater pain-related interference with
activity. Scores on the SSSQ subscale for symptoms range from 1 to 5, with higher scores indicating worse symptoms, and scores on the SSSQ
subscale for physical function range from 1 to 4, with higher scores indicating worse function. Scores on the EQ-5D scale range from 0 to 1, with
lower scores indicating worse quality of life. Scores on the PHQ-8 range from 0 to 24, with higher scores indicating more depressive symptoms.
Scores on the GAD-7 scale range from 0 to 21, with higher scores indicating more severe anxiety. The average treatment effect (ATE) the
between-group difference in the outcomes at 6 weeks is shown for each outcome.

18 n engl j med 371;1nejm.orgjuly 3, 2014

The New England Journal of Medicine

Downloaded from nejm.org on December 3, 2014. For personal use only. No other uses without permission.
Copyright 2014 Massachusetts Medical Society. All rights reserved.
 Glucocorticoid Injections for Spinal Stenosis

stenosis, there is little evidence of effectiveness (which was longer in the glucocorticoid group),
from clinical trials. Limited conclusions can be the glucocorticoidlidocaine group had signifi-
drawn from the few previous trials of epidural cantly greater improvement in the RMDQ score
glucocorticoid injections for spinal stenosis, be- at 6 weeks, but the difference was small. Pa-
cause they were small, uncontrolled, or con- tients who received glucocorticoids plus lido-
ducted without the use of fluoroscopy.29 Our caine also reported greater treatment satisfac-
large, controlled trial, in which both patients tion and greater reductions in depressive
and clinicians were unaware of the treatment symptoms. At 3 weeks, patients in the glucocor-
assignments, provides no evidence of a treat- ticoidlidocaine group had slightly more im-
ment effect at 6 weeks with fluoroscopically provement in pain and function outcomes than
guided epidural injections of glucocorticoids did patients who received lidocaine alone. This
plus lidocaine as compared with lidocaine alone. small treatment benefit observed at 3 weeks or
In analyses adjusted for the duration of pain other unmeasured glucocorticoid effects (e.g.,

Table 3. Adverse Events and Cortisol Suppression at 6 Weeks.

Lidocaine GlucocorticoidLidocaine
Adverse Event (N=200) (N=200) P Value
1 event no. of patients (%) 31 (15.5) 43 (21.5) 0.08
Total adverse events no. of events (event rate) 34 (0.17) 58 (0.29) 0.02
Adverse events according to approach no. of events/total
no. of patients (event rate)
Interlaminar approach 14/139 (0.10) 32/143 (0.22) 0.02
Transforaminal approach 20/61 (0.33) 26/57 (0.46) 0.27
Reported symptoms or events no.
Excessive pain 7 5
Headache 3 8
Fever, infection, or both 2 10
Dizziness, light-headedness, or both 4 4
Numbness, tingling, or both 4 5
Cardiovascular problems, lung problems, or both 2 6
Falls 2 4
Facial flushing 0 3
Skin irritation 4 2
Leg swelling 1 2
Dural puncture 1 1
Other 4 11
Serious adverse events: hospitalization, surgery, or both no. 4 5
Morning cortisol level no. of patients/total no. (%)*
3 wk
<3 g/dl 1/151 (0.7) 16/163 (9.8) <0.001
<10 g/dl 64/151 (42.4) 98/163 (60.1) 0.001
6 wk
<3 g/dl 0/148 (0) 5/152 (3.3) 0.06
<10 g/dl 56/148 (37.8) 81/152 (53.3) 0.01

* Because of the small number of patients, 3-week and 6-week cortisol levels below 3 g per deciliter were compared with the use of Fishers
exact test. To convert the values for cortisol to nanomoles per liter, multiply by 27.59.

n engl j med 371;1nejm.orgjuly 3, 2014 19

The New England Journal of Medicine
Downloaded from nejm.org on December 3, 2014. For personal use only. No other uses without permission.
Copyright 2014 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

reduced fatigue) may have decreased symptoms among patients who received injections that in-
of depression and increased patient satisfaction cluded glucocorticoids; these findings are consis-
with glucocorticoids. tent with systemic absorption of glucocorticoids.
In analyses stratified according to injection Patients in both treatment groups had de-
approach, transforaminal injections of glucocor- creased pain and improved function. Potential
ticoids plus lidocaine showed no significant ben- explanations for the similar improvements in the
efit over transforaminal injections of lidocaine two groups include placebo effects, regression to
alone at 3 or 6 weeks. Interlaminar injections of the mean, the natural history of spinal stenosis,
glucocorticoids plus lidocaine were associated and other factors present in both study groups,
with significant advantages over interlaminar including contact with study personnel and re-
injections of lidocaine alone with respect to ceipt of lidocaine. We did not include a sham
function and pain at 3 weeks, but the observed injection group and thus cannot assess whether
between-group differences were of minimal clin- lidocaine alone conferred a benefit. Although it
ical significance and were not apparent at 6 weeks. has been proposed that epidural lidocaine alone
Previous studies have suggested greater efficacy may have longer-term therapeutic effects,37-39
of transforaminal glucocorticoid injections than asustained benefit from lidocaine injection
of interlaminar injections.30-32 Our study was alone has not been rigorously demonstrated.
not designed to directly compare the effective- In conclusion, in the treatment of symptoms
ness of transforaminal versus interlaminar injec- of lumbar spinal stenosis, epidural injections of
tions. Because we did not randomly assign pa- glucocorticoids plus lidocaine offered minimal
tients according to injection approach, we cannot or no benefit over epidural injections of lido-
rule out the possibility that other factors associ- caine alone at 6 weeks.
ated with the injection approach contribute to or
account for the difference. Supported by a grant (1R01HS019222-01) from the Agency for
Healthcare Research and Quality.
Serious adverse events were rare in our study. Dr. Bresnahan reports holding stock in and being a former
However, there were more reported adverse employee of Johnson & Johnson. Dr. Smuck reports receiving
events in the glucocorticoidlidocaine group consulting fees from ArthroCare and grant support through his
institution from Cytonics and Seikagaku. Dr. Jarvik reports re-
than in the lidocaine-alone group and in the ceiving fees for serving on an advisory board for General Electric
transforaminal subgroup than in the interlami- and consulting fees from HealthHelp and holding patents (is-
nar subgroup. Exogenously administered gluco- sued to PhysioSonics) regarding acoustic palpation with the use
of noninvasive ultrasonographic techniques for identification of
corticoids have systemic effects that can include target sites and assessment of chronic pain disorders. No other
suppression of the hypothalamicpituitary axis33,34 potential conflict of interest relevant to this article was reported.
and reduced bone mineral density, with an in- Disclosure forms provided by the authors are available with
the full text of this article at NEJM.org.
creased risk of fracture.35,36 We observed higher We thank the LESS scientific advisory board and Pradeep
rates of cortisol suppression at 3 and 6 weeks Suri, M.D., for his contributions in the review of adverse events.

appendix
The authors affiliations are as follows: Comparative Effectiveness, Cost and Outcomes Research Center (J.L.F., B.A.C., J.A.T., P.J.H.,
S.D.S., Z.B., B.W.B., L.K., J.G.J.), and the Departments of Rehabilitation Medicine (J.L.F., J.A.T., C.S.), Biostatistics (B.A.C., P.J.H.),
Psychiatry and Behavioral Sciences (J.A.T.), Radiology (Z.B., B.W.B., J.G.J.), and Health Services (B.W.B., L.K.), University of Washing-
ton, Seattle; the Departments of Family Medicine and Internal Medicine (R.A.D.), Public Health and Preventive Medicine (R.A.D.), and
Anesthesiology and Perioperative Medicine (D.S.), and the Oregon Institute of Occupational Health Sciences (R.A.D.), Oregon Health
and Science University, and Kaiser Permanente Center for Health Research (R.A.D.) all in Portland; the Division of Research, Kaiser
Permanente Northern California, Oakland (A.L.A.), Department of Physical Medicine and Rehabilitation, Kaiser Permanente Martinez
Medical Center, Martinez (A.C.), the Department of Physical Medicine and Rehabilitation, Kaiser Permanente Redwood City Medical
Center (W.F.), and the Department of Physical Medicine and Rehabilitation, Stanford University Medical Center (D.J.K., M.S.) both
in Redwood City, Spine Care Services, and Complementary and Alternative Medicine (H.G.), and the Department of Physical Medicine
and Rehabilitation (L.W.), Kaiser Permanente San Jose Medical Center, San Jose, and the Department of Physical Medicine and Rehabili-
tation, Kaiser Permanente Roseville Medical Center, Roseville (M.T.) all in California; the Department of Anesthesiology, Perioperative
and Pain Medicine, Brigham and Womens Hospital and Spine Unit, Harvard Vanguard Medical Associates (S.S.N.), Department of Anes-
thesiology and Pain Medicine, St. Elizabeths Medical Center (F.J.G.), and Center for Pain Medicine, Massachusetts General Hospital (C.G.)
all in Boston; the Neuroscience Institute, Henry Ford Hospital (D.R.N.), and the Departments of Orthopedic Surgery (S.M.) and Radi-
ology (W.S.), Henry Ford Health System both in Detroit; the Department of Physical Medicine and Rehabilitation, University of Colo-
rado, Aurora (V.A.); the Department of Physical Medicine and Rehabilitation, Dallas Veterans Affairs Medical Center, Dallas (T.A.); the
Department of Radiology, Mayo Clinic, Rochester, MN (F.D.); and the Departments of Anesthesiology and Psychiatry, University of
Pittsburgh, Pittsburgh (A.W.).

20 n engl j med 371;1nejm.orgjuly 3, 2014

The New England Journal of Medicine

Downloaded from nejm.org on December 3, 2014. For personal use only. No other uses without permission.
Copyright 2014 Massachusetts Medical Society. All rights reserved.
 Glucocorticoid Injections for Spinal Stenosis

references
1. Harrast MA. Epidural steroid injec- response. Am J Phys Med Rehabil 2011; 28. Roland M, Fairbank J. The Roland-
tions for lumbar spinal stenosis. Curr Rev 90:1050-5. Morris Disability Questionnaire and the
Musculoskel Med 2008;1:32-8. 15. Manchikanti L, Cash KA, McManus Oswestry Disability Questionnaire. Spine
2. Deyo RA, Mirza SK, Martin BI, CD, Pampati V, Abdi S. Preliminary results (Phila Pa 1976) 2000;25:3115-24. [Erratum,
Kreuter W, Goodman DC, Jarvik JG. of a randomized, equivalence trial of fluo- Spine 2001;26:847.]
Trends, major medical complications, and roscopic caudal epidural injections in man- 29. Bresnahan BW, Rundell SD, Dagada-
charges associated with surgery for lum- aging chronic low back pain: Part 4 spi- kis MC, et al. A systematic review to as-
bar spinal stenosis in older adults. JAMA nal stenosis. Pain Physician 2008;11:833-48. sess comparative effectiveness studies in
2010;303:1259-65. 16. Ammendolia C, Stuber K, de Bruin epidural steroid injections for lumbar spi-
3. Daffner SD, Wang JC. The pathophys- LK, et al. Nonoperative treatment of lum- nal stenosis and to estimate reimburse-
iology and nonsurgical treatment of lum- bar spinal stenosis with neurogenic clau- ment amounts. PM R 2013;5:705-14.
bar spinal stenosis. Instr Course Lect dication: a systematic review. Spine (Phila 30. Schaufele MK, Hatch L, Jones W. Inter-
2009;58:657-68. Pa 1976) 2012;37(10):E609-E616. laminar versus transforaminal epidural
4. Englund J. Lumbar spinal stenosis. 17. Friedly JL, Bresnahan BW, Comstock injections for the treatment of symptom-
Curr Sports Med Rep 2007;6:50-5. B, et al. Study protocol Lumbar Epi- atic lumbar intervertebral disc hernia-
5. Friedly J, Chan L, Deyo R. Increases in dural steroid injections for Spinal Steno- tions. Pain Physician 2006;9:361-6.
lumbosacral injections in the Medicare sis (LESS): a double-blind randomized 31. Lee JH, An JH, Lee SH. Comparison of
population: 1994 to 2001. Spine (Phila Pa controlled trial of epidural steroid injec- the effectiveness of interlaminar and bi-
1976) 2007;32:1754-60. tions for lumbar spinal stenosis among lateral transforaminal epidural steroid
6. Friedly J, Nishio I, Bishop MJ, May- older adults. BMC Musculoskelet Disord injections in treatment of patients with
nard C. The relationship between repeat- 2012;13:48. lumbosacral disc herniation and spinal
ed epidural steroid injections and subse- 18. Roland M, Morris R. A study of the stenosis. Clin J Pain 2009;25:206-10.
quent opioid use and lumbar surgery. natural history of back pain. Part I: devel- 32. Thomas E, Cyteval C, Abiad L, Picot
Arch Phys Med Rehabil 2008;89:1011-5. opment of a reliable and sensitive mea- MC, Taourel P, Blotman F. Efficacy of
7. Manchikanti L, Pampati V, Boswell sure of disability in low-back pain. Spine transforaminal versus interspinous corti-
MV, Smith HS, Hirsch JA. Analysis of the (Phila Pa 1976) 1983;8:141-4. costeroid injection in discal radiculalgia
growth of epidural injections and costs in 19. Dworkin RH, Turk DC, Wyrwich KW, a prospective, randomised, double-
the Medicare population: a comparative et al. Interpreting the clinical importance blind study. Clin Rheumatol 2003;22:
evaluation of 1997, 2002, and 2006 data. of treatment outcomes in chronic pain 299-304.
Pain Physician 2010;13:199-212. clinical trials: IMMPACT recommenda- 33. Lansang MC, Farmer T, Kennedy L.
8. Manchikanti L, Pampati V, Falco FJ, tions. J Pain 2008;9:105-21. Diagnosing the unrecognized systemic
Hirsch JA. Assessment of the growth of 20. Ostelo RW, Deyo RA, Stratford P, et al. absorption of intra-articular and epidural
epidural injections in the Medicare popu- Interpreting change scores for pain and steroid injections. Endocr Pract 2009;15:
lation from 2000 to 2011. Pain Physician functional status in low back pain: towards 225-8.
2013;16(4):E349-E364. international consensus regarding mini- 34. Kay J, Findling JW, Raff H. Epidural
9. Barre L, Lutz GE, Southern D, Cooper mal important change. Spine (Phila Pa triamcinolone suppresses the pituitary-
G. Fluoroscopically guided caudal epidural 1976) 2008;33:90-4. adrenal axis in human subjects. Anesth
steroid injections for lumbar spinal steno- 21. Jordan K, Dunn KM, Lewis M, Croft P. Analg 1994;79:501-5.
sis: a retrospective evaluation of long term A minimal clinically important difference 35. Al-Shoha A, Rao DS, Schilling J, Peter-
efficacy. Pain Physician 2004;7:187-93. was derived for the Roland-Morris Dis- son E, Mandel S. Effect of epidural steroid
10. Botwin K, Brown LA, Fishman M, Rao ability Questionnaire for low back pain. injection on bone mineral density and
S. Fluoroscopically guided caudal epidural J Clin Epidemiol 2006;59:45-52. markers of bone turnover in postmeno-
steroid injections in degenerative lumbar 22. Tan G, Jensen MP, Thornby JI, Shanti pausal women. Spine (Phila Pa 1976)
spine stenosis. Pain Physician 2007;10: BF. Validation of the Brief Pain Inventory 2012;37(25):E1567-E1571.
547-58. for chronic nonmalignant pain. J Pain 36. Mandel S, Schilling J, Peterson E, Rao
11. Botwin KP, Gruber RD, Bouchlas CG, 2004;5:133-7. DS, Sanders W. A retrospective analysis of
et al. Fluoroscopically guided lumbar 23. Kroenke K, Strine TW, Spitzer RL, vertebral body fractures following epi-
transformational epidural steroid injec- Williams JB, Berry JT, Mokdad AH. The dural steroid injections. J Bone Joint Surg
tions in degenerative lumbar stenosis: an PHQ-8 as a measure of current depression Am 2013;95:961-4.
outcome study. Am J Phys Med Rehabil in the general population. J Affect Disord 37. Bicket MC, Gupta A, Brown CH IV,
2002;81:898-905. 2009;114:163-73. Cohen SP. Epidural injections for spinal
12. Briggs VG, Li W, Kaplan MS, Eskan- 24. Spitzer RL, Kroenke K, Williams JB, pain: a systematic review and meta-analy-
der MS, Franklin PD. Injection treatment Lwe B. A brief measure for assessing sis evaluating the control injections in
and back pain associated with degenera- generalized anxiety disorder: the GAD-7. randomized controlled trials. Anesthesi-
tive lumbar spinal stenosis in older adults. Arch Intern Med 2006;166:1092-7. ology 2013;119:907-31.
Pain Physician 2010;13(6):E347-E355. [Er- 25. Brazier J, Jones N, Kind P. Testing the 38. van Eijs F, Geurts J, van Kleef M, et al.
ratum, Pain Physician 2011;14:217.] validity of the Euroqol and comparing it Predictors of pain relieving response to
13. Cooper G, Lutz GE, Boachie-Adjei O, with the SF-36 health survey question- sympathetic blockade in complex region-
Lin J. Effectiveness of transforaminal naire. Qual Life Res 1993;2:169-80. al pain syndrome type 1. Anesthesiology
epidural steroid injections in patients 26. Stucki G, Daltroy L, Liang MH, Lipson 2012;116:113-21.
with degenerative lumbar scoliotic steno- SJ, Fossel AH, Katz JN. Measurement 39. Cohen SP, Kapoor SG, Rathmell JP.
sis and radiculopathy. Pain Physician 2004; properties of a self-administered outcome Intravenous infusion tests have limited
7:311-7. measure in lumbar spinal stenosis. Spine utility for selecting long-term drug ther-
14. Cosgrove JL, Bertolet M, Chase SL, (Phila Pa 1976) 1996;21:796-803. apy in patients with chronic pain: a sys-
Cosgrove GK. Epidural steroid injections 27. Bang H, Ni L, Davis CE. Assessment tematic review. Anesthesiology 2009;111:
in the treatment of lumbar spinal stenosis of blinding in clinical trials. Control Clin 416-31.
efficacy and predictability of successful Trials 2004;25:143-56. Copyright 2014 Massachusetts Medical Society.

n engl j med 371;1nejm.orgjuly 3, 2014 21

The New England Journal of Medicine
Downloaded from nejm.org on December 3, 2014. For personal use only. No other uses without permission.
Copyright 2014 Massachusetts Medical Society. All rights reserved.