(488) adverse events associated with fluoroscopically guided zygapophyseal joint injections
TRANSCRIPT
Accepted Manuscript
Adverse Events Associated With Fluoroscopically Guided LumbosacralTransforaminal Epidural Steroid Injections
Christopher Plastaras, MD, Zachary McCormick, MD, Cynthia Garvan, PhD, DonaldMacron, MD, Anand Joshi, MD, Gary Chimes, MD, Wesley Smeal, MD, JoshuaRittenberg, MD, David J. Kennedy, MD
PII: S1529-9430(15)00556-2
DOI: 10.1016/j.spinee.2015.05.034
Reference: SPINEE 56356
To appear in: The Spine Journal
Received Date: 15 April 2014
Revised Date: 3 April 2015
Accepted Date: 29 May 2015
Please cite this article as: Plastaras C, McCormick Z, Garvan C, Macron D, Joshi A, Chimes G,Smeal W, Rittenberg J, Kennedy DJ, Adverse Events Associated With Fluoroscopically GuidedLumbosacral Transforaminal Epidural Steroid Injections, The Spine Journal (2015), doi: 10.1016/j.spinee.2015.05.034.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service toour customers we are providing this early version of the manuscript. The manuscript will undergocopyediting, typesetting, and review of the resulting proof before it is published in its final form. Pleasenote that during the production process errors may be discovered which could affect the content, and alllegal disclaimers that apply to the journal pertain.
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
Fluoroscopically Guided Lumbosacral Transforaminal Epidural Steroid Injections:
Adverse Events and Predictive Factors
Christopher Plastaras,1 Zachary McCormick,2 Cynthia Garvan,3 Donald Macron,4 Anand
Joshi,5 Gary Chimes,6 Wesley Smeal,7 Joshua Rittenberg8 ,David J. Kennedy9
1. Christopher Plastaras, MD.
Department of PM&R, University of Pennsylvania, Philadelphia, PA.
2. Zachary McCormick, MD.
Department of PM&R, Northwestern McGaw Medical Center/The Rehabilitation
Institute of Chicago, Chicago, IL.
3. Cynthia Garvan, PhD.
College of Nursing, University of Florida, Gainesville, FL.
4. Anand Joshi, MD.
Department of Orthopaedics, Duke University Medical Center, Durham, NC.
5. Donald Macron, MD.
Department of Neurological Surgery, Stony Brook Neurosciences Institute, Stony
Brook, NY.
6. Gary Chimes MD.
Lake Washington Sports and Spine. Bellevue, WA.
7. Wesley Smeal, MD.
Alegent Health Physical Medicine and Rehabilitation Clinic, Omaha, NE.
8. Joshua Rittenberg, MD.
Kaiser Permanente Department of Physical Medicine and Rehabilitation, Oakland,
CA
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
9. David J. Kennedy, MD.
Stanford University Department of Orthopeadics, Palo Alto, CA
Corresponding Author:
Zack McCormick, MD
Department of Physical Medicine and Rehabilitation
Rehabilitation Institute of Chicago
McGaw Medical Center, Northwestern University Feinberg School of Medicine
Chicago, Illinois
Email: [email protected]
Phone: 510-388-7084
Fax: N/A
Conflict of Interest Statement: None of the authors report any financial conflicts of interest.
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
1
Adverse Events Associated With Fluoroscopically Guided Lumbosacral 1
Transforaminal Epidural Steroid Injections 2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
2
Abstract 1
2
Background Context: While the types and incidence of adverse events associated with 3
transforaminal epidural steroid injection (TFESI) have been described, no study has used 4
a systematic standardized questionnaire to solicit adverse events (AEs) from patients in 5
order to capture an accurate range and incidence of complications. 6
Purpose: To systematically identify the types and incidence of AEs associated with 7
TFESI. Additionally, to evaluate demographic and clinical factors that may predict a 8
higher risk of an AE. 9
Study Design/Setting: A retrospective cohort study from a multi-physician academic 10
PM&R clinic. 11
Patient Sample: Patients, ages 19-89, who underwent a fluoroscopically-guided TFESI 12
for lumbosacral radicular pain between 2004-2007 were included. 13
Outcome Measures: The relationship of adverse events with gender, age, trainee 14
presence, steroid type, pre-procedure visual analogue scale (VAS) pain score, systolic 15
blood pressure, fluoroscopy time and corticosteroid injectate volume were analyzed. 16
Methods: AE data were collected using a survey both immediately and at 24-72 hours 17
following TFESI. Statistical analysis was performed using chi-square, Fisher’s exact or 18
Wilcoxon rank sum two-sided tests. Logistic regression analysis was also performed. 19
C.P. is the owner of RICPLAS computer software. 20
Results: In 1,295 consecutive patients undergoing 2,025 TFESI procedures, immediate 21
AEs and delayed AEs occurred after 182 (9.2%) and 305 (20.0%) injections, respectively. 22
The most common immediate AEs were: vasovagal reaction (4.2%) and interrupted 23
procedure from intravascular flow (1.7%). Common delayed AEs included: pain 24
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
3
exacerbation (5.0%), injection site soreness (3.9%), headache (3.9%), facial 1
flushing/sweating (1.8%) and insomnia (1.6%). Significant associations were identified 2
between adverse events and gender, age, pre-procedure VAS, steroid type and 3
fluoroscopy time. Trainee involvement in the procedure did not impact the complication 4
rate. 5
Conclusions: 6
Fluoroscopically-guided lumbosacral TFESI is associated with a similar rate of minor 7
adverse events both immediately and 24-72 hours post-procedure that are typical of other 8
axial corticosteroid injections. Permanent adverse events were not found in this sample. 9
The most common adverse events associated with TFESI include vasovagal episodes, 10
procedure interruption from intravascular flow, pain exacerbation, injection site soreness, 11
headache, and insomnia. 12
13
Key Words: fluoroscopic injection, transforaminal, epidural, complications, adverse 14
effects, steroid injection 15
Level of Evidence: Level II16
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
4
Introduction 1
2
Lumbosacral radicular pain is commonly treated with epidural steroid injection.1 3
Evidence is more robust for the efficacy of the transforaminal epidural injection approach 4
compared to other routes of epidural steroid deposition in the treatment of lumbosacral 5
radicular pain.2-10 Complications reported in association with transforaminal epidural 6
steroid injection (TFESI) include major events such as epidural or subdural hematoma, 7
epidural abscess, discitis and paraplegia (Kennedy/Dreyfuss, MacVicar 2013),8,11 as well 8
as minor temporary adverse events such as vasovagal reaction, pain exacerbation, 9
headache and facial flushing.12-15 Only Karaman and colleagues have studied the rate of 10
adverse events prospectively. In a cohort of 562 patients, these investigators found no 11
major complications and an 11.5% rate of minor adverse events.13 This study design 12
suffered from the lack of a standardized questionnaire or protocol to solicit adverse 13
events from patients. Consequently, this study did not capture the true range and 14
incidence of minor adverse events accurately, as only a limited range of minor adverse 15
events that are known to occur with corticosteroid injections were reported. Thus, the 16
goal of the present study was to systematically identify the types and incidence of adverse 17
events associated with TFESI. Additionally, this study evaluated demographic and 18
clinical factors associated with these adverse events with the goal of providing clinicians 19
with valuable information that will help predict which patients are at higher risk of 20
experiencing an adverse event. 21
22
Methods 23
24
The Northwestern University Institution Review Board approved this retrospective cohort 25
study. Consecutive subjects were identified through electronic medical record review of 26
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
5
all individuals seen at an urban, academic, physical medicine and rehabilitation outpatient 1
musculoskeletal and spine center. The study included all individuals, age 18-89 who 2
underwent at least one lumbosacral TFESI at this facility between March 8, 2004 and 3
April 19, 2007. There were no exclusion criteria. Physicians board certified in Physical 4
Medicine and Rehabilitation with additional subspecialty board certification in either 5
Pain Medicine or Sports Medicine ordered and performed all injections. A trainee in 6
Physical Medicine and Rehabilitation residency or Sports Medicine fellowship performed 7
injections in 47% of cases, as the study was performed at an academic teaching facility. 8
The attending physicians described above supervised and/or intervened upon these 9
injections in order to ensure adequate technique. The indication for a lumbosacral TFESI 10
in this practice was a clinical history consistent with radicular pain, corroborated with 11
MRI findings of radiographic pathology at the level identified clinically. While there is 12
an inherent false positive rate of symptomatic pathology associated with MRI imaging of 13
the lumbar spine,16,17 corroboration of clinical symptoms and MRI findings represents the 14
current standard for diagnosing radicular pain.18 An injection was performed only if an 15
individual met these criteria, failed to respond to conservative treatment, and still had 16
significant functional limitations. The specific level chosen for injection was based 17
primarily on clinical history with imaging as supportive, not independently definitive, 18
additional data. 19
20
All injections were performed using the subpedicular transforaminal technique.19 The 21
patient was placed in a prone position, and the skin was prepped with sterile technique. 22
The fluoroscope was positioned appropriately in order to provide an oblique view of the 23
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
6
subpedicular space. The skin and soft tissue were anesthetized using 1% lidocaine 1
(preservative-free). A sterile 22 gauge 3.5, 5, or 7 inch spinal needle was then advanced 2
to the superior aspect of the neural foramen above the exiting spinal nerve. Anterior-3
posterior, oblique, and lateral fluoroscopic views were obtained to confirm accurate 4
needle placement. Subsequently, 1-2 mL of Isovue 300 contrast was injected through 5
microbore tubing under live fluoroscopy. If intravenous uptake occurred, the needle was 6
repositioned until intravenous uptake was absent and an epidural flow pattern was 7
achieved. If intra-arterial, intrathecal, or intradiscal flow was identified, the procedure 8
was aborted. Then, 1.5 - 2 mL of 1% lidocaine was injected as an anesthetic test dose, 9
and if no adverse events were noted after waiting 1-2 minutes, 1-2 mL of corticosteroid 10
(betamethasone 6mg/mL or triamcinolone 40mg/mL) was administered through 11
microbore tubing. Volumes and doses depended on the number of sites injected. For 12
unilateral single level procedures, 2 mL of 1% lidocaine and 2 mL of steroid were used. 13
For unilateral, two level procedures or bilateral uni-level procedures, 1.5 mL of lidocaine 14
and 1 mL of steroid were used at each site. Bilateral injections were performed for 15
patients with bilateral symptoms. Two level unilateral injections were performed for 16
patients with multilevel disease and/or for those who had not received adequate pain 17
relief with a unilateral one level TFESI. “Multilevel disease” indicates that a patient had 18
clinical symptoms consistent with radicular pain at more than one 19
dermatomal/sclerotomal spinal nerve root level, corroborated by radiographic pathology 20
at multiple corresponding spinal levels. No patients in this cohort had injections at more 21
than two sites during a single visit in order to maximize diagnostic specificity and 22
minimize risks associated with increased table time and multiple injection levels.15 23
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
7
1
Patient demographic information was collected prior to the procedure. Adverse event data 2
was collected immediately post-procedure. To assure accuracy, the treating physician 3
entered all procedural data immediately via a drop-down menu and free text in the 4
electronic clinical database. Additional adverse events could also be entered into the 5
clinical database immediately by nursing staff in the post-procedure recovery area. 6
Immediate pain relief was determined by a 50% or greater decrease in pain on a visual 7
analogue scale (VAS). 8
9
For all delayed adverse event data, a nurse uninvolved with data analysis conducted 10
standard follow-up telephone calls at 24-72 hours after the procedure, and entered data 11
via drop-down menu choices and free text immediately into the clinical database. 12
Adverse events reported during or immediately after the procedure, but before the patient 13
was discharged from the visit were designated “immediate adverse events.” Events 14
reported during the follow-up telephone call were denoted “delayed adverse events.” A 15
standardized questionnaire (Table 1) was administered by trained nursing staff during 16
follow-up telephone calls in order to expedite identification of potentially serious 17
complications requiring immediate evaluation (e.g., progressive weakness) and in order 18
to provide reassurance regarding common symptoms after TFESI (e.g., facial flushing). 19
This questionnaire was adapted from Botwin et al.’s study of complications.14 Individuals 20
were also asked about symptoms not specifically addressed by the questionnaire in a free 21
text field designated for “other comments.” Individuals were excluded from analysis if 22
they could not be contacted within 24-72 hours by telephone. 23
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
8
1
Clinical and adverse event data were recorded in a discrete structured clinical database 2
(RICPLAS© - Rehabilitation Institute of Chicago Physiatric Log & Analysis System) 3
(Tables 2 and 3). This database was used as a routine method of clinical documentation 4
that was subsequently entered into the hospital medical record. De-identified data were 5
obtained from this database through queries designed in Microsoft Access 2003 and 6
Microsoft SQL Server 2000. Three senior researchers (CP, WS, JR) coded this data 7
independently. For example, an adverse event was coded as a “vagal episode” if “vagal 8
episode” was documented by a clinician or if symptoms and signs consistent with vagal 9
symptoms were documented, such as diaphoresis, dizziness, and relative bradycardia or 10
hypotension. These were still classified as vagal events even if symptoms occurred 11
without frank syncope. Discrepancies in independent coding were reconciled by 12
consensus decision between the three senior researchers. The coded data was entered 13
into Microsoft Excel 2003. 14
15
Statistical Analysis 16
SAS version 9.3 (Cary, N.C) was used for all data analysis. Two-sided tests were used for 17
all hypothesis testing and a level of significant was set at 0.05. In order to investigate 18
demographic and clinical factors associated with adverse events, data from each patient’s 19
first procedure was included in the analysis. This was done in order to assure 20
independence of observations and valid statistical results. Chi-square or Fisher’s exact 21
tests were used to determine the relationship between adverse events and categorical 22
variables (e.g. steroid type). Wilcoxon rank sum tests were used to determine the 23
relationship between adverse events and numerical variables (e.g., age). Additional 24
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
9
analysis was performed in order to investigate demographic and clinical factors 1
associated with the immediate and delayed adverse events that occurred at a minimum 2
rate of 1%; these included vasovagal episode, intravascular flow, delayed pain 3
exacerbation, injection site soreness, facial flushing, headache, and insomnia. Logistic 4
regression analysis was performed to simultaneously investigate which predictive factors 5
were most related to the adverse events of interest. 6
7
Results 8
One thousand two hundred and ninety five consecutive patients including 577 men 9
(44.6%) and 718 women (55.4%) underwent lumbosacral TFESIs. A total of 2,025 10
procedures were performed (Tables 4). The mean (SD) age of individuals included was 11
52.8 (16.2) years, ranging from 19 to 89. Patient clinical and procedural details are 12
displayed in Table 5. 13
14
The incidence of immediate and delayed adverse events after TFESI are shown in Tables 15
6 and 7. Immediate and delayed adverse events occurred after 182 (9.2%) and 305 16
(20.0%) injections, respectively. Only two immediate adverse events occurred in greater 17
than 1% of procedures: vasovagal episode (4.2%) and intravascular flow that changed or 18
interrupted the procedure (1.7%). Delayed adverse events that occurred in at least 1% of 19
patients included pain exacerbation (5.0%), injection site soreness (3.9%), headache 20
(3.9%), facial flushing/sweating (1.8%) and insomnia (1.6%). Five patients (0.3%) visited 21
an emergency room or were hospitalized. The reasons for an emergency room visit or 22
hospitalization included: low back pain without leg symptoms found to have no new 23
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
10
pathology (n=3), self-limited dizziness in the context of a cardiac history (n=1), and 1
gastroenteritis (n=1). No pain exacerbations were associated with neurologic change. 2
3
The relationship of demographic, clinical, and procedural factors to immediate and 4
delayed adverse events after TFESI is shown in Table 8. In the results of the logistic 5
regression analysis, the following relationships remained significant: immediate adverse 6
events were associated with greater duration of fluoroscopy time (p < 0.001); delayed 7
adverse events were associated with younger age (p < 0.01) and female gender (p < 0.01). 8
9
The relationship of demographic, clinical, and procedural factors to adverse events that 10
occurred at a frequency of greater than 1% after TFESI is shown in Table 9. In the results 11
of the logistic regression analysis, the following relationships remained significant: 12
vasovagal episodes were associated with younger age (p < 0.001), male gender (p < 0.01) 13
and lower pre-procedure VAS pain score (p < 0.001); intravascular flow that changed or 14
interrupted the procedure was associated with greater duration of fluoroscopy time (p < 15
0.001) and higher pre-procedure VAS pain score (p < 0.01); the most common delayed 16
adverse event (pain exacerbation) occurred more often in younger patients (p < 0.001); 17
injection site soreness was more common in younger patients (p < 0.01) and when 18
triamcinolone as opposed to betamethasone was used (p < 0.01); facial flushing was more 19
frequent in women (p < 0.05); headaches occurred more commonly in women (p < 0.05) 20
and when betamethasone as opposed to triamcinolone was used (p < 0.01). 21
22
Discussion 23
24
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
11
In this cohort of 1,295 consecutive patients who underwent 2,025 TFESIs, serious 1
complications (ie neurologic compromise) did not occur. Minor complications with no 2
long-term sequelae including intradiscal flow (0.4%), temporary motor or sensory spinal 3
block (0.4%), and dural puncture (0.2%) occurred at a frequency similar to that described 4
in other studies.12,20-22 Only transient, expected side-effects of corticosteroid injections 5
occurred at a frequency of greater than 1%. 6
The present data suggest that transient minor immediate adverse events occur more 7
frequently with TFESI (9.2%) compared to other axial injections, including that of the 8
sacroiliac (SI) joint (2.6%)23 and the zyagaphophyseal (ZA) joint (4.1%).24 Similar to the 9
studies of SI and ZA joint injections, these data show that vasovagal episodes are the 10
most common immediate adverse event during TFESI, and this reaction occurs at a 11
somewhat greater frequency (5.0%) compared to injections of the SI (2.1%)23 and ZA 12
(3.7%)24 joints. This observed rate of vasovagal reactions after TFESI is similar to that 13
reported for lumbar medial branch blocks (5.1%)25 and falls within the range of other 14
studies that have defined the incidence of this complication after TFESI (0.3%-15
8.7%).13,14,26 16
17
These data indicate that delayed minor transient adverse events occur at a similar 18
frequency with TFESI (20.0%) compared to other axial injections, including that of the 19
SI joint (24.2%)23 and the ZA joint (18.9%).24 As in studies of SI and ZA joint 20
injections, the current data show that pain exacerbations are the most common immediate 21
adverse event after TFESI and that this occurs at a similar frequency after TFESI (5.0%) 22
as compared to injections of the SI (5.3%)23 and ZA (4.3%)24 joints. Injection site 23
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
12
soreness (3.9%), and facial flushing or sweating (1.8%) after TFESI occurred less 1
commonly or at a similar frequency compared with SI (12.9% and 2.3%, respectively) 2
and ZA (6.0% and 1.1%, respectively) joint injections.23,24 Insomnia after TFESI (1.6%) 3
occurred at a similar rate compared to ZA joint injections (2.2%).24 Transient headache 4
(3.8%) occurred more commonly with TFESI than with SI (0.8%) or ZA (1.6%) 5
injections.23,24 6
7
Vasovagal episodes, in addition to being more common in younger patients, were more 8
likely to occur in males and in those with a lower pre-procedure VAS pain score. It is 9
unclear why lower pre-procedure pain scores are associated with a greater risk of 10
vasovagal reaction. While, greater pain might be associated with a higher baseline blood 11
pressure and pulse that could potentially compensate for the physiologic reductions in 12
these vital signs during a vasovagal episode, we did not find significant differences in 13
baseline blood pressure or pulse in those who experienced a vagal episode versus those 14
who did not in this cohort. Regardless, clinicians should be aware of younger age, male 15
gender, and lower pre-procedure pain levels as risk factors for a vasovagal episode. 16
17
Analysis revealed additional relationships between adverse events, age, and gender. 18
Younger age and female gender were associated with the occurrence of delayed adverse 19
events. Younger individuals were more likely to report delayed pain exacerbation and 20
injection site soreness. Women were more likely to experience facial flushing/sweating 21
and transient headaches. Thus, younger age and female gender appear to be risk factors 22
for a number of delayed adverse events. 23
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
13
1
With regards to steroid choice, particulate steroids were used exclusively in this cohort. 2
Injection site soreness was more often reported after TFESI with triamcinolone compared 3
to betamethasone. Headache was more common with the use of betamethasone compared 4
to triamcinolone. These findings should be considered when selecting the type of 5
corticosteroid for individual patients. 6
7
As expected, intravascular flow that changed or interrupted the procedure was associated 8
with longer fluoroscopy time. 9
10
The strengths of this study include a large sample, a standardized data collection 11
instrument used for the purpose of future retrospective data analysis and practice audit, 12
and consensus among multiple physicians when categorizing adverse events from chart 13
review. 14
15
There are several limitations to this study. The data presented reflect complications 16
associated with routine practice at one academic spine center; thus, these findings 17
primarily apply to similar practice patterns. Bias may exist, as the treating physician was 18
partially responsible for entering immediate adverse event data. However, there were no 19
incentives for the physician to hide any adverse event data, and an independent observer 20
(nursing staff) uninvolved in data-analysis collected all of the delayed adverse event data. 21
Additionally, this cohort included all consecutive patients to further decrease the risk of 22
bias. Lastly, the follow-up period of 24-72 hours post-procedure was brief. This follow-23
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
14
up period may have been too short to capture all delayed complications, such as 1
infectious or endocrine-related adverse effects of corticosteroid injection. It should, 2
however, have captured all significant complications related to hematoma or neurologic 3
infarction. Future study would ideally include follow-up at 1 month, which has been 4
pursued in other multicenter collaborative efforts.27 Multiple testing was performed, 5
which may have increased experiment wise error, another possible limitation. 6
7
8
Conclusions 9
Fluoroscopically-guided lumbosacral TFESI is associated with a similar rate of minor 10
adverse events both immediately and 24-72 hours post-procedure that are typical of other 11
axial corticosteroid injections. Permanent adverse events were not found in this sample. 12
Common events included: vasovagal episodes, intravascular flow that changed or 13
interrupted the procedure, pain exacerbation, injection site soreness, transient headache, 14
facial flushing/sweating, and insomnia. Clinicians should be aware that male patients are 15
more likely to experience vasovagal episodes; female patients are more likely to 16
experience delayed adverse events, particularly headache or facial flushing; younger 17
individuals are more likely to experience a vasovagal episode, a delayed pain 18
exacerbation or injection site soreness. Injection site soreness occurs more often with 19
triamcinolone compared to betamethasone, while the inverse relationship is true with 20
regards to headache. 21
22
23
24
References 25
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
15
1
1. Friedly J, Chan L, Deyo R. Increases in lumbosacral injections in the Medicare 2
population: 1994 to 2001. Spine. 2007 Jul 15. 32(16):1754-1760. 3
2. Roberts ST, Willick SE, Rho ME, Rittenberg JD. Efficacy of lumbosacral 4
transforaminal epidural steroid injections: a systematic review. PMR. 2009 5
Jul;1(7):657-68. 6
3. Boswell M, Hansen H, Trescot A, Hirsch J. Epidural steroids in the management 7
of chronic spinal pain and radiculopathy. Pain Physician. 2003; 6: 319–334. 8
4. Abdi S, Datta S, Trescot A, Schultz D, Adlaka R, Atluri S, Smith 9
HS, Manchikanti L. Epidural steroids in the management of chronic spinal pain: 10
A systematic review. Pain Physician. 2007; 10: 185–212. 11
5. Vad VB, Bhat AL, Lutz GE, Cammisa F. 12
Transforaminal epidural steroid injections in lumbosacral radiculopathy: 13
a prospective randomizedstudy. Spine (Phila Pa 1976). 2002 Jan 1;27(1):11-6. 14
6. Riew KD, Yin Y, Gilula L, Bridwell KH, Lenke LG, Lauryssen C, Goette K. 15
The effect of nerve-root injections on 16
the need for operative treatment of lumbar radicular pain. A 17
prospective, randomized, controlled, double-blind study. J Bone Joint Surg 18
Am. 2000 Nov;82-A(11):1589-93. 19
7. Ghahreman A, Ferch R, Bogduk N. The Efficacy of Transforaminal Injection of 20
Steroids for the Treatment of Lumbar Radicular Pain. Pain Medicine 2010; 11: 21
1149-1168. 22
8. MacVicar J, King W, Landers MH, Bogduk N. The effectiveness of lumbar 23
transforaminal injection of steroids: a comprehensive review with systematic 24
analysis of the published data. Pain Med. 2013 Jan;14(1):14-28. 25
9. Kennedy, David J., Christopher Plastaras, Ellen Casey, Christopher J. Visco, 26
Joshua D. Rittenberg, Bryan Conrad, James Sigler, and Paul Dreyfuss. 27
Comparative Effectiveness of Lumbar Transforaminal Epidural Steroid Injections 28
with Particulate Versus Nonparticulate Corticosteroids for Lumbar Radicular Pain 29
due to Intervertebral Disc Herniation: A Prospective, Randomized, Double-Blind 30
Trial. Pain Medicine. 2014 Apr;15(4):548-55. 31
10. McCormick Z, Plastaras C. Lumbosacral transforaminal epidural steroid 32
injections are equally effective for treatment of lumbosacral radicular pain in the 33
obese compared to non-obese population: a pilot study. J Back Musculoskelet 34
Rehabil. 2013 Jan 1;26(2):183-8. 35
11. Kennedy DJ, Dreyfuss P, April CN, and Bogduk N. Paraplegia Following Image-36
Guided Transforaminal Lumbar Spine Epidural Steroid Injection: Two Case 37
Reports. Pain Med. 2009; 10(8): 1389–1394. 38
12. McGrath JM, Schaefer MP, Malkamaki DM. Incidence and characteristics of 39
complications from epidural steroid injections. Pain Med. 2011 May;12(5):726-40
31. 41
13. Karaman H, Kavak GO, Tüfek A, Yldrm ZB. The complications of 42
transforaminal lumbar epidural steroid injections. Spine (Phila Pa 1976). 2011 43
Jun;36(13):E819-824. 44
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
16
14. Botwin KP, Gruber RD, Bouchlas CG, Torres-Ramos FM, Freeman TL, Slaten 1
WK. Complications of fluoroscopically guided transforaminal lumbar epidural 2
injections. Arch Phys Med Rehabil. 2000 Aug; 81(8):1045-1050. 3
15. Stalcup ST, Crall TS, Gilula L, Riew KD. Influence of needle-tip position on the 4
incidence of immediate complications in 2,217 selective lumbar nerve root 5
blocks. The Spine Journal 6.2 (2006): 170-176. 6
16. Jensen MC, Brant-Zawadzki MN, Obuchowski N, Modic MT, Malkasian D, Ross 7
JS. Magnetic resonance imaging of the lumbar spine in people without back pain. 8
N Engl J Med. 1994 Jul 14;331(2):69-73. 9
17. Boden SD, Davis DO, Dina TS, Patronas NJ, Wiesel SW. Abnormal magnetic-10
resonance scans of the lumbar spine in asymptomatic subjects. A prospective 11
investigation. J Bone Joint Surg Am. 1990 Mar;72(3):403-8. 12
18. Kreiner DS, Hwang SW, Easa JE, Resnick DK, Baisden JL, Bess S, Cho 13
CH, DePalma MJ, Dougherty P 2nd, Fernand R, Ghiselli G, Hanna AS, Lamer 14
T, Lisi AJ, Mazanec DJ, Meagher RJ, Nucci RC, Patel RD, Sembrano JN, Sharma 15
AK, Summers JT, Taleghani CK, Tontz WL Jr,Toton JF; North American Spine 16
Society. An evidence-based clinical guideline for 17
the diagnosis and treatment of lumbar disc herniation with radiculopathy. Spine 18
J. 2014 Jan;14(1):180-91. 19
19. Bogduk, N. International Spinal Injection Society Guidelines for the Performance 20
of Spinal Injection Procedures. Part 1: Zygapophysial Joint Blocks. The Clinical 21
Journal of Pain. 1997; 13(4): 285–302. 22
20. Candido KD, Katz JA, Chinthagada M, McCarthy RA, Knezevic NN. Incidence 23
of intradiscal injection during lumbar fluoroscopically guided transforaminal and 24
interlaminar epidural steroid injections. Anesth Analg. 2010 May 1;110(5):1464-25
7. 26
21. Plastaras CT, Casey E, Goodman BS, Chou L, Roth D, Rittenberg J. 27
Inadvertent intradiscal contrast flow during lumbar transforaminal epidural 28
steroid injections: a case series examining the prevalence 29
of intradiscal injection as well as potential associated factors and adverse events. 30
Pain Med. 2010 Dec;11(12):1765-73. 31
22. Lee HI, Park YS, Cho TG, Park SW, Kwon JT, Kim YB. Transient adverse 32
neurologic effects of spinal pain blocks. J Korean Neurosurg Soc. 2012 33
Sep;52(3):228-33. 34
23. Plastaras CT, Joshi AB, Garvan C, Chimes GP, Smeal W, Rittenberg J, Lento 35
P, Stanos S, Fitzgerald C. Adverse events associated with fluoroscopically 36
guided sacroiliac joint injections. PMR. 2012 Jul;4(7):473-478. 37
24. Plastaras C, McCormick Z, Macron D, Garvan C, Joshi A, Chimes G, Smeal W, 38
Rittenberg J. Adverse Events Associated with Fluoscopically Guided 39
Zygapophyseal Joint Injections. Pain Physician. Accepted for publication April 40
2014. 41
25. Kennedy DJ, Schneider B, Casey E, Rittenberg J, Conrad B, Smuck M, Plastaras 42
CT. Vasovagal Rates in Flouroscopically Guided Interventional Procedures: A 43
Study of Over 8,000 Injections. Pain Med. 2013 Dec;14(12):1854-1859. 44
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
17
26. Trentman TL, Rosenfeld DM, Seamans DP, Hentz JG, Stanek JP. Vasovagal 1
reactions and other complications of cervical vs. lumbar translaminar epidural 2
steroid injections. Pain Pract. 2009 Jan-Feb;9(1):59-64. 3
27. Schneider BJ, Smuck M, Plastaras C, Nahm LS, Maus TP, Kennedy DJ. Delayed 4
Complications Following Interventional Pain Procedures. Poster 458 for the 5
American Academy of Physical Medicine and Rehabilitation Annual Assembly 6
and Technical Exhibition in San Diego, CA October 2014. PM&R Journal 6, no. 7
9 (2014): S346. 8
9
10
11
12
13
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
Table 1: Follow Up Telephone Questionnaire
Headache? Yes No Explanation:_______________________________
Fevers? Yes No Explanation:_______________________________
Rash? Yes No Explanation:_______________________________
Injection site swelling? Yes No Explanation:_______________________________
Increased pain? Yes No Explanation:_______________________________
New/worsening weakness? Yes No Explanation:_______________________________
Pain Rating? ___
Do you need to make a follow-up appointment? Yes No
Questions for the doctor? ____________________________________________________
Other comments: ___________________________________________________________
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
Table 2: List of Adverse Events Monitored During and Immediately Post Procedure
Increased Pain that changed or interrupted procedure
Vagal
• Vagal—Injection Completed
• Vagal—Injection Discontinued
Cardiovascular Instability
• HTN (symptomatic)
• Tachycardia that changed or interrupted procedure
Sensorimotor Block (Motor Block)
Allergic Reaction to Medication
Contrast Abnormality/Technical
• Technical —required alternate location or approach, completed
• Technical—could not position at target, not completed
Patient-related
• Patient movement/positioning that changed or interrupted procedure
• Anxiety/Patient request to stop that changed or interrupted procedure
Others
• Nausea without bradycardia or hypotension (i.e., not vasovagal)
• Dizziness/lightheadedness without bradycardia or hypotension (i.e., not
vasovagal)
• Shakiness that changed or interrupted procedure
• Elevated, but asymptomatic BP that lead to discontinuation of procedure
• Diaphoresis without nausea or change in vital signs (i.e., not vasovagal)
Notables
• Steroid particle stopping flow through needle
• Hiccups
• Steroid clogged needle
• Face burning
• Throat fullness
• Arm numbness
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
Table 3: List of Delayed Adverse Events Monitored at Follow-up Telephone Call
Headache – transient
Headache – severe
Fever
Chills
Rash
Facial Flushing/sweating
Injection Site Swelling
Pain exacerbation
Weak
Numbness
Cramping
Pressure
Spasms
Injection Site Soreness
Nausea/Vomiting
Diarrhea
Bowel incontinence
Sleeplessness
Mood fluctuation/Anxiety/Depressed/Crying
“Shakiness”/”Worked Up”/”Jittery”
Dizziness/Lightheaded
Vasovagal reaction
Hiccups
Sneezing
Elevated blood sugar
Flu-like symptoms
Ear filled with fluid feeling
Fatigue
“Cold sensation in hands and feet”
Elevated blood pressure
Hospitalization/ER Visit
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
Table 4: Patient Description
Mean (SD)
or
n (%) reported
Number of Patients 1,295
Number of Injections 2,025
Age 52.8 (16.2)
Gender
Male
Female
-
577 (44.6%)
718 (55.4%)
Radiologic Diagnosis
Herniated Nucleus Pulposus
Radicular Pain without
definitive radiologic
correlation
Central Stenosis
Degenerative Disc Disease
Spondylolisthesis
Foraminal Stenosis
409 (31.6%)
309 (23.9%)
225 (17.4%)
169 (13.1%)
92 (7.1%)
55 (4.3%)
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
Table 5: Patient Clinical and Procedural Details
Mean (SD)
or
n (%) reported
Pre-procedure VAS 5.8 (2.2)
Immediate Post-procedure VAS 3.7 (2.4)
Systolic BP 129.7 (16.6)
Diastolic BP 80.5 (9.3)
Pulse 80.2 (13.7)
Fluoroscopy time (seconds) 35.6 (21.6)
Administration
Trainee involved
Attending only
612 (47.3%)
683 (52.7%)
Level Injected
L1-L2
L2-L3
L3-L4
L4-L5
L5-S1
S1
12 (0.59%)
39 (1.93%)
142 (7.02%)
441 (21.79%)
1002 (49.41%)
388 (19.17%)
Side of Procedure
Right
Left
Bilateral
553 (42.7%)
519 (40.1%)
222 (17.2%)
Number of Procedures
1
2
3
4
5
6
7
Total
786 (60.7%)
344 (26.6%)
130 (10.0%)
21 (1.6%)
10 (0.8%)
1 (0.1%)
3 (0.2%)
2,025
Intravascular flow that changed
or interrupted the procedure
No
1991 (98.3%)
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
Yes 34 (1.7%)
BP: Blood pressure
VAS: Visual analogue pain scale
Table 6: Immediate Adverse Events after TFESI
Immediate Adverse Event
n (%)
Number of procedures with data available 2,025
Total Adverse Events 186 (9.2%)
Vasovagal episode 85 (5.0%)
Vasovagal episode—injection completed 59 (2.9%)
Intravascular flow that changed or interrupted the
procedure
34 (1.7%)
Vasovagal episode – Injection discontinued 26 (2.1%)
Increased pain that changed or interrupted the
procedure
17 (0.8%)
Alternate injection approach or location required –
injection completed
17 (0.8%)
Intradiscal flow pattern 7 (0.4%)
Dizziness/lightheadedness without bradycardia or
hypotension
6 (0.3%)
Patient movement/positioning that changed or
interrupted procedure
5 (0.3%)
Temporary motor-spinal block 4 (0.2%)
Dural Puncture 3 (0.2%)
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
Nausea without bradycardia or hypotension 3 (0.2%)
Alternate injection approach or location required –
injection not completed
2 (0.1%)
Temporary sensory-spinal block 2 (0.1%)
Steroid-clogged needle 2 (0.1%)
Sensation of facial burning 1 (<0.1%)
Sensation of throat fullness 1 (<0.1%)
Symptomatic hypertension 1 (<0.1%)
Table 7: Delayed Adverse Events after TFESI
Delayed Adverse Event
n (%) reported
Number of procedures with data available 1,523
Total Adverse Events 305 (20.0%)
Pain exacerbation 76 (5.0%)
Injection site soreness 59 (3.9%)
Headache - transient 58 (3.8%)
Facial flushing/sweating 28 (1.8%)
Insomnia 25 (1.6%)
Injection site swelling 13 (0.9%)
Fever 13 (0.9%)
Nausea/vomiting 10 (0.7%)
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
Rash 9 (0.6%)
Sensation of pressure at injection site 7 (0.5%)
Mood fluctuation/anxiety/depression 6 (0.4%)
Subjective weakness 6 (0.4%)
Cramping 5 (0.3%)
Numbness 5 (0.3%)
Elevated Blood Sugar 5 (0.3%)
Hospitalization/Emergency Room visit 5 (0.3%)
Hypertension 4 (0.3%)
Subjective shakiness 4 (0.3%)
Hiccups 4 (0.3%)
Fatigue 3 (0.2%)
Spasms 3 (0.2%)
Dizziness/lightheadedness 3 (0.2%)
Chills 2 (0.1%)
Diarrhea 2 (0.1%)
Flu-like symptoms 2 (0.1%)
Sneezing 1 (<0.1%)
Vasovagal episode 1 (<0.1%)
Cold sensation in hands and feet 1 ( <0.1%)
Bowel incontinence 1 (<0.1%)
Headache - severe 1 (<0.1%)
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
Table 8: Relationship of Demographic, Clinical, and Procedural Factors to Adverse Events.
Bolded numbers indicate associations that remained significant after logistic regression
analysis.
Immediate Adverse Event
Mean (SD) or n (%) reported
Delayed
Adverse Event
Mean (SD) or n (%) reported
Gender
Male
Female
67 (11.6%)*
60 (8.4%)
62 (14.7%)
131 (23.3%)***
Administration
Trainee involved
Attending only
65 (10.7%)
62 (9.0%)
94 (20.3%)
99 (19.7%)
Steroid type
Betamethasone, AE
Triamcinolone, AE
41 (7.1%)
46 (6.8%)
88 (21.6)%
101 (18.3%)
Age
No AE
Yes AE
53.0 (16.3)
51.4 (15.6)
54.1 (16.6)
50.3 (15.8)**
Pre-procedure VAS
No AE
Yes AE
5.9 (2.2)
5.8 (2.3)
5.9 (2.2)
6.0 (2.2)
Systolic BP
No AE
Yes AE
128.7 (16.5)
128.7 (17.2)
129.3 (17.0)
127.8 (16.6)
Diastolic BP
No AE
Yes AE
80.5 (9.2)
81.1 (9.5)
80.4 (9.4)
80.3 (9.1)
Pulse
No AE
Yes AE
80.4 (13.7)
79.4 (12.9)
80.8 (13.6)
81.3 (13.4)
Pulse Oximetry (% Hgb O2
Sat)
No AE
Yes AE
96.7 (7.7)
97.1 (2.3)
96.7 (8.0)
97.2 (1.4)
Fluoroscopy time
(seconds)
No AE
Yes AE
34.6 (20.4)
51.6 (33.7)***
36.4 (23.5)
34.7 (21.0)
Level Injected: Yes AE
L1-L2
L2-L3
L3-L4
L4-L5
0 (0.0)
1 (4.2)
8 (8.9)
25 (9.8)
2 (100.0)
3 (17.7)
18 (27.7)*
41 (19.8)
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
L5-S1
S1
62 (9.4)
30 (11.6)
102 (20.2)
27 (14.3)
Foraminal Stenosis:
No Foraminal Stenosis –
Yes AE
Yes Foraminal Stenosis
- Yes AE
116 (9.4)
11 (19.3)*
182 (19.4)
11 (25.0)
Immediate Pain Relief
No Relief - Yes AE
Yes Relief – Yes AE
58 (11.5)
58 (7.6)*
65 (17.0)
125 (21.3)
AE: Adverse event
BP: Blood pressure
SD: Standard deviation
VAS: Visual analogue pain scale
% Hgb O2 Sat: percent hemoglobin oxygen saturation
* p < 0.05
** p < 0.01
*** p < 0.001
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
Table 9: Relationship Demographic, Clinical, and Procedural Factors to Adverse Events that Occur at a Frequency of Greater
than 1%. Frequencies and percentages reported are listed for categorical variables and means (standard deviations) are listed
for numerical variables. Bolded numbers indicate associations that remained significant after logistic regression analysis.
Vasovagal
Episode
Intravascular
Flow
Delayed
Pain
Exacerbation
Injection Site
Soreness
Facial
Flushing
Headache Insomnia
Gender
Male, AE
Female, AE
40 (6.9%)**
24 (3.3%)
7 (1.2%)
10 (1.4%)
15 (3.6%)
34 (6.1%)
11 (2.6%)
28 (5.0%)
2 (0.5%)
15 (2.7%)*
10 (2.4%)
30 (5.3%)*
5 (1.2%)
10 (1.8%)
Administration
Trainee, AE
Attending, AE
36 (5.9%)
28 (4.1%)
7 (1.2%)
10 (1.5%)
24 (5.2%)
25 (4.8%)
18 (3.9%)
21 (4.0%)
12 (2.6%)*
5 (1.0%)*
20 (4.3%)
20 (3.8)%
6 (1.3%)
9 (1.7%)
Steroid type
Betamethasone, AE
Triamcinolone, AE
28 (4.9%)
23 (3.4%)
5 (0.9%)
4 (0.6%)
26 (2.7%)
23 (2.4%)
7 (1.7%)
31 (5.6%)**
8 (2.0%)
9 (1.6%)
24 (5.9%)**
14 (2.5%)
9 (2.2%)
6 (1.1%)
Age
No AE
Yes AE
53.1 (16.3)
46.1 (12.6)***
52.7 (16.2)
55.9 (16.3)
53.7 (16.4)
45.1 (15.0)***
53.6 (16.5)
45.6 (15.1)**
53.2 (16.5)
60.6 (11.8)
52.7 (16.2)
55.1 (15.1)
53.3 (16.5)
52.9 (17.0)
Pre-procedure VAS
No AE
Yes AE
5.9 (2.2)
5.1 (2.1)**
5.9 (2.2)
7.4 (1.9)**
5.9 (2.2)
6.2 (2.4)
5.9 (2.2)
5.4 (2.3)
5.9 (2.2)
5.6 (1.4)
5.9 ± 2.2
6.3 ± 2.1
5.9 (2.2)
4.9 (2.0)
Systolic BP
No AE
Yes AE
128.9 (16.5)
125.9 (17.9)
128.8 (16.6)
126.9 (14.5)
129.3 (17.0)
122.8 (13.4)**
129.0 (16.9)
129.0 (17.9)
128.9 (16.9)
132.7 (16.7)
128.7 (16.6)
128.3 (17.6)
129.0 ± 16.9
129.8 ± 21.0
Diastolic BP
No AE
Yes AE
80.5 (9.2)
81.6 (9.4)
80.6 (9.2)
78.3 (9.0)
80.5 (9.4)
78.1 (8.0)
80.4 (9.3)
79.2 (9.7)
80.3 (9.4)
82.8 (7.0)
80.5 (9.2)
80.6 (9.4)
80.3 (9.4)
83.4 (6.9)
Pulse
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
No AE
Yes AE
80.4 (13.7)
78.2 (11.6)
80.3 (13.6)
75.1 (13.2)
80.7 (13.5)
84.2 (14.5)
80.9 (13.6)
82.2 (11.9)
80.9 (13.5)
81.8 (13.1)
80.4 (13.7)
82.0 (12.9)
80.9 (13.5)
79.9 (13.0)
Pulse Oximetry
(%Hgb O2 Sat)
No AE
Yes AE
96.7 (7.6)
97.3 (1.2)
96.7 (7.4)
97.3 (1.2)
96.7 (7.4)
97.4 (1.2)
96.7 (7.4)
97.2 (1.6)
96.8 (7.3)
96.8 (1.5)
96.7 (7.5)
97.0 (1.7)
96.8 (7.3)
97.0 (1.3)
Fluoroscopy time
No AE
Yes AE
36.3 (22.8)
35.2 (18.2)
35.6 (21.4)
85.6 (45.3)***
36.0 (23.1)
37.3 (22.1)
36.2 (22.9)
33.1 (25.5)
36.2 (23.1)
29.8 (16.6)
36.4 (22.7)
34.4 (17.3)
36.1 23.1)
37.2 (18.6)
Volume of
Corticosteroid (mL)
No AE
Yes AE
1.8 (0.4)
1.8 (0.4)
1.8 (0.4)
1.4 (0.5)*
1.8 (0.4)
1.9 (0.3)
1.8 (0.4)
1.8 (0.4)
1.8 (0.4)
1.7 (0.5)
1.8 (0.4)
1.8 (0.5)
1.8 (0.4)
1.9 (2.6)
Level Injected: Yes AE
L1-L2
L2-L3
L3-L4
L4-L5
L5-S1
S1
0 (0.0)
0 (0.0)
3 (3.3)
12 (4.7)
34 (5.2)
15 (5.8)
0 (0.0)
0 (0.0)
2 (2.2)
2 (0.8)
6 (0.9)
7 (2.7)
0 (0.0)
1 (5.9)
4 (6.2)
6 (2.9)
31 (6.2)
7 (3.7)
0 (0.0)
0 (0.0)
3 (4.6)
7 (3.4)
24 (4.8)
5 (2.7)
0 (0.0)
0 (0.0)
3 (4.6)*
8 (3.9)
6 (1.2)
0 (0.0)
0 (0.0)
3 (17.7)*
6 (9.2)
5 (2.4)
17 (3.4)
9 (4.8)
0 (0.0)
2 (11.8)**
3 (4.6)
1 (0.5)
7 (1.4)
2 (1.1)
Foraminal Stenosis:
No Foraminal
Stenosis – Yes AE
Yes Foraminal
Stenosis - Yes AE
60 (4.9)
4 (7.0)
16 (1.3)
1 (1.8)
48 (5.1)
1 (2.3)
36 (3.8)
3 (6.8)
15 (1.6)
2 (4.6)
39 (4.2)
1 (2.3)
15 (1.6)
0 (0.0)
Immediate Pain Relief
No Relief - Yes AE
Yes Relief – Yes AE
30 (6.0)
30 (4.0)
7 (1.4)
10 (1.3)
16 (4.2)
33 (5.6)
14 (3.7)
23 (3.9)
6 (1.6)
11 (1.9)
6 (1.6)
34 (5.8)**
6 (1.6)
9 (1.5)
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
AE: Adverse event
BP: Blood pressure
VAS: Visual analogue pain scale
% Hgb O2 Sat: percent hemoglobin oxygen saturation
* p < 0.05
** p < 0.01
*** p < 0.001