ABSTRACT

Objectives. Cervical transforaminal epidural injections of corticosteroids have been used in the treatment of radicular pain. Particulate agents have been associated with rare adverse neurological outcomes. It is unknown whether nonparticulate preparations are any less effective than particulate preparations. Therefore, a study was designed to determine whether there is a basis for promoting a theoretically safer nonparticulate corticosteroid preparation.

Design. Volunteer patients were randomized to receive a single cervical transforaminal epidural injection with one of two corticosteroid preparations.

Setting. This study was undertaken in a private practice setting.

Patients. Those with single-level, unilateral radicular pain with advanced imaging demonstrating single-level neural compression.

Interventions. Patients received a single cervical transforaminal epidural injection with either dexamethasone or triamcinolone.

Outcome Measures. Ratings were obtained by an independent unbiased assessor at 4 weeks via a telephone interview. A visual analog pain scale was used preprocedurally and a verbal integer scale was used at 4 weeks to assess the severity of the patient's radicular pain. As a secondary outcome measure, a patient-specified functional outcome measure was obtained.

Results. Both groups exhibited statistically and clinically significant improvements in pain at 4 weeks. Although the triamcinolone group exhibited a somewhat greater improvement, the difference between groups was not significantly different.

Conclusion. The study found that the effectiveness of dexamethasone was slightly less than that of triamcinolone, but the difference was neither statistically nor clinically significant. A theoretically safer nonparticulant agent appears to be a valid alternative to particulate agents that have been used to date, and which have been associated with hazard.

Introduction

No treatment has been validated for the treatment of cervical radicular pain. Conservative therapy, in the form of physiotherapy, traction, analgesics, collars, and exercises, is frequently recommended, but observational studies report conflicting results. Good outcomes have been claimed in as many as 80% of patients [1,2], or as few as 40%[3,4]. Meanwhile, controlled trials have shown that these interventions provide no attributable effect beyond that of placebo or natural history [5–7]. By reputation, surgery is the mainstay of treatment [8,9]; but the one controlled study found it to be no more effective than physiotherapy or wearing a collar at 12-month follow-up [10].

In an effort to fill this therapeutic vacuum, some practitioners have advocated transforaminal injection of corticosteroids. The rationale for this intervention is that radicular pain is caused by inflammation of a spinal nerve or its roots [11,12], and that accurate injection of corticosteroids directly onto the affected nerve should suppress that inflammation and relieve the pain.

Cervical transforaminal injection of corticosteroids has been advocated solely on the basis of descriptive studies, all of which utilized particulate corticosteroids. Bush and Hillier [13] reported that 76% of their 68 patients achieved complete relief of arm pain, but these investigators used three types of treatment, and it is not possible from their report to determine what proportion of patients responded explicitly to transforaminal injections. Slipman et al. [14] reported good or excellent results in 60% of 20 patients, followed for 12–45 months, after an average of 2.2 injections. Of the 32 patients described by Vallee et al. [15], 29% had complete relief of pain at 3 months. This proportion persisted at 6 months, but diminished to 20% at 12 months. An additional 29% of patients obtained at least 50% relief of their pain at 3 months. This proportion persisted at 6 months, but fell to 18% at 12 months.

These studies paint an encouraging picture of the effectiveness of cervical transforaminal injections of corticosteroids, particularly in the light of there being no other validated alternative. To date, however, they have not been vindicated by controlled trials. Nevertheless, this lack of formal evidence has not prevented the popularization of cervical transforaminal injections.

Meanwhile, serious complications have occurred with the use of this treatment. These include cortical blindness [16], quadriplegia, and death as a result of infarction of the spinal cord or brainstem [17–19]. Numerous other cases are reportedly still subjudice [20,21]. These complications have been attributed to inadvertent injection either into either the vertebral artery [16–18] or into a radicular artery that reinforces the anterior spinal artery [18–21].

Inadvertent injection into a vertebral artery is an avoidable complication. It occurs if incorrect technique is used, and if arterial injection is not recognized when a test dose of contrast medium is administered. Injection into a radicular artery is substantially harder to detect due to the fleetingly brief flow of contrast medium in this smaller caliber vessel. Moreover, injection into a radicular artery can occur even with accurate needle placement [20].

If injection into a radicular artery is not recognized and particulate steroids are subsequently injected, spinal cord infarction could occur if the particles act as an embolus [20,21]. An anatomical study showed that the size of particles in commonly used steroid preparations equals or exceeds the caliber of many radicular arteries [18]. A case report described anesthetization of the spinal cord following injection of local anesthetic, but not steroids, into a cervical intervertebral foramen [22]. Together, these reports suggest that the use of particulate steroids constitutes a major risk factor for cervical transforaminal injections.

In principle, embolization should not occur when nonparticulate steroids are used. However, despite this theoretical advantage, particulate steroids have traditionally been utilized for transforaminal injections, despite the availability of nonparticulate preparations [13–15].

The present study was undertaken to determine whether nonparticulate steroid preparations are any less effective than particulate preparations. Specifically, the purpose was to determine whether there might be a basis for promoting a theoretically safer agent, either for the conduct of controlled trials or for conventional practice. For logistic reasons, it was not planned as an efficacy study. Problems in recruiting sufficient volunteer patients to consent to randomization precluded a placebo-controlled trial with multiple arms.

Methods

The study was conducted in a private practice in Seattle, Washington, USA, specializing in the diagnosis and treatment of spine pain. The recruitment period was between January 26, 2004, and September 2, 2004. Four hundred and twenty patients were referred in this time frame for possible cervical epidural steroid injections.

Thirty consecutive patients who met the inclusion and exclusion criteria and consented to participate were enrolled. The inclusion criteria were: radicular pain associated with unilateral nerve root compression at a single segmental level, as seen on computed tomography or magnetic resonance imaging; pain in the upper limb greater than in the neck; no litigation; no worker's compensation or disability remuneration; and the willingness to complete a phone interview 4 weeks after treatment. The exclusion criteria were: other pain or a comorbid condition that might interfere with assessment of relief of the radicular pain and central spinal stenosis less than 8 mm.

All patients were treated with a single injection performed according to the protocol recommended by the International Spine Intervention Society [23]. Under fluoroscopic guidance a 25-gauge Quinke needle (Kimberly-Clark Ballard Medical Products, Draper, UT, USA) was placed in the intervertebral foramen of the target nerve, against the posterior wall of that foramen, opposite its midpoint. A test dose of contrast medium was administered in order to check for obvious intravascular injection. Injection of contrast medium using real-time fluoroscopy and digital subtraction angiography was then utilized to exclude more subtle intravascular uptake. In the absence of intravascular injection, 0.75–1.0 mL of lidocaine 4% was injected. Subsequently, the assigned corticosteroid preparation was injected. Patients were randomly allocated to receive either a nonparticulate preparation (12.5 mg dexamethasone sodium phosphate: Decadron, Sicor Pharmaceuticals Inc, Irvine, CA, USA), or a particulate preparation (60 mg triamcinolone acetonide: Kenalog, Bristol-Myers Squibb Co, Princton, NJ, USA).

A pragmatic outcome measure was used. A visual analog pain scale was used preprocedurally. A verbal integer scale was used at the 4-week telephone interview. Patients were asked to rate the severity of their radicular pain. Ratings were obtained by an independent unbiased assessor. From these data, the percentage reduction of pain was calculated, which became the primary outcome measure.

As a secondary outcome measure, a patient-specified functional outcome was used [24–26]. At inception, patients were asked to nominate four activities of daily living that they felt were limited by their pain, and which most dearly they would wish restored. At follow-up they were asked which and how many of these activities they had regained. Additionally, work status was assessed before and after treatment.

To compare pain scores, within and between groups, a t-test was used. In order to compare the proportions of patients who obtained particular degrees of improvement the 95% confidence intervals (CI) of these proportions were calculated. For the correlation between improvement in pain and restoration of activities of daily living a Pearson correlation coefficient was calculated.

Results

The two groups did not differ significantly in demographic features or clinical features at inception (Table 1). The women in the triamcinolone group were younger than the women in the dexamethasone group, but their ages were not significantly different from the ages of the men in either group. No complications occurred following any of the injections.

Table 1

Demographic and clinical features of patients treated for cervical radicular pain with transforaminal injections of either dexamethasone or triamcinalone

Feature Group
 
P Value 
Dexamethasone Triamcinolone 
Gender, N 
  Male  9  7 0.464 
  Female  6  8  
Age (years), mean ± SD (range) 
  Male 47 ± 14.6 (25–73) 50 ± 7.9 (42–58) 0.615 
  Female 55 ± 6.2 (44–60) 47 ± 6.9 (36–54) 0.033 
Duration of pain, N 
  0–3 months  8 10 0.597 
  3–6 months  1  3  
  6–12 months  1  1  
  >12 months  3  1  
VAS (0–100), mean ± SD (range) 48 ± 22 (10–80) 49 ± 21 (20–90) 0.933 
Segmental level affected, N 
  C5  1  1 0.747 
  C6  7  5  
  C7  7  9  
Work status, N 
  Full time 12 11 0.224 
  Part time  0  2  
  Not working  1  2  
  Retired  2  0  
Feature Group
 
P Value 
Dexamethasone Triamcinolone 
Gender, N 
  Male  9  7 0.464 
  Female  6  8  
Age (years), mean ± SD (range) 
  Male 47 ± 14.6 (25–73) 50 ± 7.9 (42–58) 0.615 
  Female 55 ± 6.2 (44–60) 47 ± 6.9 (36–54) 0.033 
Duration of pain, N 
  0–3 months  8 10 0.597 
  3–6 months  1  3  
  6–12 months  1  1  
  >12 months  3  1  
VAS (0–100), mean ± SD (range) 48 ± 22 (10–80) 49 ± 21 (20–90) 0.933 
Segmental level affected, N 
  C5  1  1 0.747 
  C6  7  5  
  C7  7  9  
Work status, N 
  Full time 12 11 0.224 
  Part time  0  2  
  Not working  1  2  
  Retired  2  0  

VAS = visual analog pain scale.

Both groups exhibited statistically and clinically significant improvements in pain at 4 weeks (Table 2). Although the triamcinolone group exhibited a somewhat greater improvement, the difference between groups at follow-up was not significantly different. Within each group, the responses of patients with chronic pain (>3 months) did not differ significantly in either magnitude or distribution from those of patients with acute pain.

Table 2

Mean visual analog pain scores (0–100) and (standard deviation) of all patients treated by cervical transforaminal injection of dexamethasone or triamcinolone

 Dexamethasone Trimacinolone Between 
Baseline 48 (22) 49 (21) P = 0.933 
4 weeks 29 (26) 17 (14) P = 0.156 
Within P = 0.006 P = 0.000  
 Dexamethasone Trimacinolone Between 
Baseline 48 (22) 49 (21) P = 0.933 
4 weeks 29 (26) 17 (14) P = 0.156 
Within P = 0.006 P = 0.000  

A greater proportion of the dexamethasone group (27%) obtained complete relief of their pain than in the triamcinolone group (7%), but this difference was not statistically significant (Table 3). In the dexamethasone group the proportion of patients who obtained at least 50% relief of their pain was 0.60 (95% CI: 0.35–0.85) In the triamcinalone group, the proportion was 0.67 (95% CI: 0.43–0.91). Given the overlap of the confidence intervals, these proportions are not significantly different statistically.

Table 3

The numbers (N), proportions (Pr), and cumulative proportions (ΣPr) of patients who obtained the percentage improvements of pain (ΔVAS%) indicated after treatment with cervical transforaminal injection of either dexamethasone or triamcinolone

ΔVAS% Dexamethasone
 
Trimacinolone
 
Pr ΣPr Pr ΣPr 
100 0.27 0.27 0.07 0.07 
 90 0.00 0.27 0.00 0.07 
 80 0.00 0.27 0.40 0.47 
 70 0.07 0.34 0.07 0.54 
 60 0.07 0.41 0.07 0.60 
 50 0.20 0.60 0.07 0.67 
 40 0.00 0.60 0.07 0.74 
 30 0.07 0.67 0.13 0.87 
 20 0.00 0.67 0.00 0.87 
 10 0.13 0.80 0.00 0.87 
  0 0.07 0.87 0.13 1.00 
Worse 0.13 1.00 0.00 1.00 
ΔVAS% Dexamethasone
 
Trimacinolone
 
Pr ΣPr Pr ΣPr 
100 0.27 0.27 0.07 0.07 
 90 0.00 0.27 0.00 0.07 
 80 0.00 0.27 0.40 0.47 
 70 0.07 0.34 0.07 0.54 
 60 0.07 0.41 0.07 0.60 
 50 0.20 0.60 0.07 0.67 
 40 0.00 0.60 0.07 0.74 
 30 0.07 0.67 0.13 0.87 
 20 0.00 0.67 0.00 0.87 
 10 0.13 0.80 0.00 0.87 
  0 0.07 0.87 0.13 1.00 
Worse 0.13 1.00 0.00 1.00 

In the triamcinolone group, relief of pain was significantly and strongly correlated with restoration of activities of daily living (P = 0.000; Pearson correlation coefficient: 0.795). Of the 10 patients who obtained greater than 50% relief of their pain, seven regained all four of their desired activities of daily living; one patient regained three activities and two regained two activities. Such a relationship was lacking in the dexamethasone group (P = 0.283; Pearson correlation coefficient: 0.297), because two patients who obtained little or no relief paradoxically regained three activities, and only three of the nine patients who obtained greater than 50% relief of pain regained all four desired activities. Nevertheless, five patients regained three activities, and one regained two activities.

Discussion

It was not the intention of the present study to assess long-term effectiveness of cervical transforaminal injection of corticosteroids. Thus, extended follow-up was purposefully not performed.

The express intention of the study was to determine whether using a nonparticulate steroid preparation initially appeared to be any less effective than a commonly used particulate preparation. In that regard, the study found that the yield of dexamethasone was slightly less than that of triamcinalone, but the difference was neither statistically nor clinically significant.

Because of the small sample size, the present study lacks power (7%). Consequently, the observed lack of difference might have arisen by chance. However, to prove beyond doubt that dexamethasone is no less effective than triamcinalone would require a prohibitively large number of patients (N = 769). Even a sample size of 100 would have only 17% power. Under these circumstances, the pragmatic way to confirm the conclusions ventured would be for others to repeat the experiment.

In that regard, the present study provides nominal, objective data upon which operators can base an option. A shadow of suspicion has been cast over particulate steroids, in that they carry a risk of embolization if injected into a radicular artery. Theoretically, nonparticulate steroids do not carry that risk. In light of the present results, operators might choose to use nonparticulate steroids as a measure toward avoiding the risk of embolization.

Conclusion

It is our belief that the slightly lesser yield of dexamethasone is worth the theoretical reduction in risk of spinal cord injury, stroke, or death.

In rendering this conclusion, we do not seek to promote the use of cervical transforaminal injections. Whether or not practitioners should perform this intervention, in the absence of a controlled trial, is a matter judged on multiple other grounds. Those include, but are not limited to: the tension between offering patients a treatment that could work instead of waiting for science to catch up with patients' needs and contrasting the putative risks of transforaminal injections with those of surgery.

Rather, what we seek to contribute is solely the observation that a theoretically safer agent appears to be a valid alternative to the agents that have been used to date, and which have been associated with hazard. That observation should be of relevance both to conventional practitioners and to investigators who might plan a controlled cervical transforaminal injection study.

References

1
Honet
JC
Puri
K
.
Cervical radiculitis: Treatment and results in 82 patients
.
Arch Phys Med Rehabil
 
1976
;
57
:
12
6
.
2
Saal
JS
Saal
JA
Yurth
EF
.
Nonoperative management of herniated cervical intervertebral disc with radiculopathy
.
Spine
 
1996
;
21
:
1877
83
.
3
Martin
GM
Corbin
KB
.
An evaluation of conservative treatment for patients with cervical disk syndrome
.
Arch Phys Med Rehabil
 
1954
;
35
:
87
92
.
4
Arnasson
O
Carlsson
CA
Pellettieri
L
.
Surgical and conservative treatment of cervical spondylotic radiculopathy and myelopathy
.
Acta Neurochir
 
1987
;
84
:
48
53
.
5
British Association of Physical Medicine.
Pain in the neck and arm: A multicentre trial of the effects of physiotherapy
.
BMJ
 
1966
;
1
:
253
8
.
6
Goldie
I
Landquist
A
.
Evaluation of the effects of different forms of physiotherapy in cervical pain
.
Scand J Rehab Med
 
1970
;
2–3
:
117
21
.
7
Klaber Moffett
JA
Hughes
GI
Griffiths
P
.
An investigation of the effects of cervical traction. Part 1: Clinical effectiveness
.
Clin Rehabil
 
1990
;
4
:
205
11
.
8
Chestnut
RM
Abitbol
JJ
Garfin
SR
.
Surgical management of cervical radiculopathy
.
Orthop Clin North Am
 
1992
;
23
:
461
74
.
9
Ahlgren
BR
Garfin
SR
.
Cervical radiculopathy
.
Orthop Clin North Am
 
1996
;
27
:
253
63
.
10
Persson
LCG
Carlsson
CA
Carlsson
JY
.
Long-lasting cervical radicular pain managed with surgery, physiotherapy or a cervical collar: A prospective, randomized study
.
Spine
 
1997
;
22
:
751
8
.
11
Kang
JD
Georgescu
HI
McIntyre-Larkin
L
Stanovic-Racic
M
Evans
CH
.
Herniated cervical intervertebral discs spontaneously produce matrix metalloproteinases, nitric oxide, interleukin-6 and prostaglandin E2
.
Spine
 
1995
;
22
:
2373
8
.
12
Furusawa
N
Baba
H
Miyoshi
N
et al
Herniation of cervical intervertebral disc. Immunohistochemical examination and measurement of nitric oxide production
.
Spine
 
2001
;
26
:
1110
6
.
13
Bush
K
Hillier
S
.
Outcome of cervical radiculopathy treated with periradicular/epidural corticosteroid injections: A prospective study with independent clinical review
.
Eur Spine J
 
1996
;
5
:
319
25
.
14
Slipman
CW
Lipetz
JS
Jackson
HB
Rogers
DP
Vresilovic
EJ
.
Therapeutic selective nerve root block in the nonsurgical treatment of atraumatic cervical spondylotic radicular pain: A retrospective analysis with independent clinical review
.
Arch Phys Med Rehabil
 
2000
;
81
:
741
6
.
15
Vallee
JN
Feydy
A
Carlier
RY
Mutschler
C
Mompoint
D
Vallee
CA
.
Chronic cervical radiculopathy: Lateral approach periradicular corticosteroid injection
.
Radiology
 
2001
;
218
:
886
92
.
16
McMillan
MR
Crumpton
C
.
Cortical blindness and neurologic injury complicating cervical transforaminal injection for cervical radiculopathy
.
Anesthesiology
 
2003
;
99
:
509
11
.
17
Rozin
L
Rozin
R
Koehler
SA
et al
Death during a transforaminal epidural steroid nerve root block (C7) due to perforation of the left vertebral artery
.
Am J Forensic Med Path
 
2003
;
24
:
351
5
.
18
Tiso
RL
Cutler
T
Catania
JA
Whalen
K
.
Adverse central nervous system sequelae after selective transforaminal block: The role of corticosteroids
.
Spine J
 
2004
;
4
:
468
74
.
19
Brouwers
PJAM
Kottnik
EJBL
Simon
MAM
Prevo
RL
.
A cervical anterior spinal artery syndrome after diagnostic blockade of the right C6-nerve root
.
Pain
 
2001
;
91
:
397
9
.
20
Baker
R
Dreyfuss
P
Mercer
S
Bogduk
N
.
Cervical transforaminal injection of corticosteroids into a radicular artery: A possible mechanism for spinal cord injury
.
Pain
 
2002
;
103
:
211
5
.
21
Rathmell
JR
Aprill
C
Bogduk
N
.
Cervical transforaminal injection of steroids
.
Anesthesiology
 
2004
;
100
:
1595
600
.
22
Karasek
M
Bogduk
N
.
Temporary neurologic deficit after cervical transforaminal injection of local anesthetic
.
Pain Med
 
2004
;
5
:
202
5
.
23
International Spine Intervention Society.
Cervical transforaminal injection of corticosteroids
. In:
Bogduk
N
, ed.
Practice Guidelines for Spinal Diagnostic and Treatment Procedures.
 
San Francisco, CA
:
International Spine Intervention Society
;
2004
:
237
48
.
24
Stratford
P
Gill
C
Westaway
M
Binkley
J
.
Assessing disability and change on individual patients: A report of a patient specific measure
.
Physiother Can
 
1995
;
47
:
258
63
.
25
Westaway
MN
Stratford
PW
Binkley
JM
.
The patient-specific functional scale: Validation of its use in persons with neck dysfunction
.
J Orthop Sports Phys Ther
 
1998
;
27
:
331
8
.
26
Chatman
AB
Hyams
SP
Neel
JM
et al
The patient-specific functional scale: Measurement properties in patients with knee dysfunction
.
Phys Ther
 
1997
;
77
:
82
9
.