Abstract

Background: Despite the lack of FDA (Food and Drug Administration) approval, cervical and lumbar epidural spinal injections are frequently performed in the US to address back pain and/or painful radiculopathy. The three major types of injections include; interlaminar/translaminar (ESI), transforaminal (TFESI), or caudal injections. Notably, most studies document little to no clear short-term, and no long-term benefits/efficacy for these injections vs. various placebos.

Methods: More adverse events (AE) occurred with cervical© rather than lumbar (L) injections, and more severe AE were attributed to C-TFESI vs. CESI injections.

Results: Acute post injection AE symptoms were observed immediately or within 72 post-injection hours. These symptoms included; hypotension, acute respiratory distress, chest pain, upper extremity numbness, weakness, paresthesias, paralysis, and fevers. More AE were attributed to cervical C-TFESI vs. cervical CESI. These AE included; intramedullary/cord injections, intravascular injections (i.e. vertebral artery) resulting in brain stem/cerebellar/cord strokes, epidural abscess/infection, confusion, epidural hematomas, intracranial hypotension, and/or 6^th nerve cranial palsies. AE for lumbar LESI/L-TFESI included; infections/abscess, epidural hematomas/subdural hematomas, intravascular injections, cerebrospinal fluid (CSF) leaks/dural tears (DT), and intracranial/postural hypotension. Notably, the vast majority of studies showed little to no short-term, and no long-term benefits for cervical or lumbar ESI/TFESI vs placebos (i.e. mostly consisting of normal saline alone, or saline plus local anesthesia).

Conclusion: Epidural cervical and lumbar ESI or TFESI spinal injections demonstrated minimal to no short-term, and no long-term benefits for the treatment of cervical and/or lumbar pain/radiculopathy vs. placebos. Further, more AE were observed for cervical vs. lumbar epidural injections overall, with more AE usually seen with TFESI vs. ESI procedures.

Keywords: Spinal Epidural Injections: Interlaminar, Translaminar (ESI), Transforaminal (TFESI), Caudal: Cervical, Lumbar, Adverse Events, Cord Injections, Cerebrospinal Fluid (CSF) Leaks, Neurological Deficits, Paralysis, Vascular Injections

INTRODUCTION

Despite the lack of FDA (Food and Drug Administration) approval, Medicare reported that over 9 million cervical and lumbar epidural spinal injections were performed in 2012 [ Table 1 ].[ 11 ] The three major type of epidural injections included; interlaminar/translaminar (ESI), transforaminal (TFESI), or caudal injections.[ 1 - 22 ] More adverse events (AE) were reported for cervical vs. lumbar injections overall, and most major cervical AE (i.e., including intramedullary/cord and/or intravascular injections including stroke and death) were attributed to TFESI vs. ESI procedures. Further, many studies documented minimal/no short-term, and no long-term benefits for any of these injections (i.e. comparing epidural injections vs. placebos).

Table 1:

Summary of articles regarding spinal epidural injections.

AE Attributed to ESI/TFESI That Are Not FDA (Food and Drug Administration) Approved

Two articles discussed the lack of FDA approval of cervical or lumbar spinal epidural injections.[ 6 , 16 ] In 2013, Epstein noted that both cervical and lumbar spinal ESI/TFESI were not FDA approved, and posed significant risks of AE [ Table 1 ].[ 6 ] That article: “...cite(d) contaminated epidural steroid injections resulting in meningitis, stroke, paralysis, and death. The Center for Disease Control (CDC) specifically identified 25 deaths (many due to Aspergillosis), 337 patients sickened, and 14,000 exposed to contaminated steroids”. Further, Manchikanti et al. in 2015 recounted the FDA warning regarding the greater risks observed for cervical TFESI that included 131 major neurological AE events, including death, and 41 instances of adhesive arachnoiditis.[ 16 ]

Time of Onset of Acute/Symptoms/Signs After Cervical/Lumbar ESI/TFESI Injections

Schreiber et al. in 2016 reported that adverse symptoms/signs occurred immediately to within 72 hours following epidural injections; these typically included hypotension, acute respiratory distress, chest pain, upper extremity numbness, weakness, paresthesias, and/or paralysis, and fever [ Table 1 ].[ 20 ]

Minimal or No Short-Term and No Long-Term Benefits of Cervical ESI/TFESI

Two studies documented minimal/no short-term, and no long-term benefits of cervical ESI/TFESI vs. placebos [ Table 1 ].[ 2 , 12 ] In Anderberg et al. (2007), 40 consecutive patients with unilateral 1-2 level cervical radiculopathy were randomized to receive either TFESI/local anesthesia vs. NS (Normal Saline)/local anesthesia; at 3 post-treatment weeks, both groups demonstrated comparable outcomes. [ 2 ] In 2018, Epstein noted that cervical C-TFESI demonstrated minimal to no short-term, and no long-term benefits, but posed significant risks of AE.[ 12 ] These included; epidural hematomas, infections (abscess/meningitis), new neurological deficits (i.e. including intramedullary cord injections), intravascular injuries (brain stem, cerebellar), and strokes (i.e., largely attributed to particulate steroid emboli to distal arterial branches).[ 12 ]

Minimal to No Short-Term and No Long-Term Benefits of Lumbar ESI, TFESI, or Caudal Injections vs. Placebos

Five studies documented minimal/no short-term, and no long-term benefits of lumbar ESI/TFESI [ Table 1 ].[ 3 , 4 , 18 , 19 , 22 ] Using Methylprednisolone Acetate/Normal Saline (NS) vs NS alone (placebo) to treat 158 patients with sciatica/lumbar disc disease, Carette et al. (1997) showed only transient short-term (i.e., 6 weeks to 3 mos. post-injection).[ 4 ] Further, 1 year later, both groups had similar requirements for surgery: 25.8% with ESI vs. 24.8% without ESI. Comparing 42 control group patients (CG without steroids) vs. 43 steroid group (SG) patients with sciatica, Valat et al. (2003) found no short or long term benefits utilizing ESI (i.e. SG employing 2 ml Prednisolone Acetate (50 mg) and performing 3 ESI at 2-day intervals) vs. 1 mg NS (CG); they both demonstrated comparable frequencies of treatment failures defined by the need to administer NSAIDS (nonsteroidal anti-inflammatory agents) at 20 post-injection days.[ 22 ] In a randomized controlled study, Arden et al. (2005) utilized 3 ESI (Triamcinolone Acetonide) vs. 3 NS epidural injections given every 3 weeks to treat unilateral sciatica (i.e. patients followed for 1-18 months); those receiving ESI exhibited transient benefits up to 3 weeks following the 1^st injection, but encountered no long-term benefits beyond 6-52 weeks after ESI.[ 3 ] Parr et al. (2009) also found no short ( 6 mos.) for lumbar ESI (i.e. utilizing interlaminar, TFESI, or caudal injections) vs. placebo addressing stenosis/radiculopathy; their review of the literature yielded similar results.[ 19 ] In 2020, Oliveira et al. summarized data from 25 clinical trials comparing the efficacy of lumbar ESI (i.e. interlaminar TFESI or caudal injections) vs. subcutaneous, vs. intramuscular vs. interspinous steroid injections vs. placebo (i.e. inert, Normal Saline, local anesthesia) in the treatment of 2470 patients with lumbar radiculopathy/sciatica. [ 18 ] Over the 4 periods they reviewed (i.e. 2 weeks to 3 mos. (short-term); > 3 mos.=< 12 mos (intermediate term); and >/= 12 mos. (long-term)), ESI vs. placebo showed only small short-term benefits that were “not... considered clinically important”, and there were no long-term benefits.

More Minor vs. Major AE Reported for Cervical and/or Lumbar ESI, TFESI, or Caudal Lumbar Injections

Five studies documented more minor vs. major AE attributed to cervical and/or lumbar epidural injections; additionally some reports stated more AE occurred following TFESI vs. ESI [ Table 1 ].[ 1 , 5 , 15 , 16 , 17 ] When Abbasi et al. (2007) reviewed the literature from multiple databases, they found a 0-16.8% incidence of AE for interlaminar CESI (0-16.8%); most AE were minor.[ 1 ] When McGrath et al. (2011) compared the incidence of AE in 1857 patients treated over 7 years with 3964 lumbar TFESI vs. 161 cervical ESI, 123 lumbar ESI, and 17 caudal injections, they observed 103 minor AE (2.4% incidence per injections). [ 17 ] These included; increases pain (1.1%), pain at injection sites (0.33%), persistent numbness 0.14%), and other factors (0.8%).[ 17 ] They concluded that fewer AE occurred following TFESI (2.1%) vs. ESI (6%). Over a 5-year period, Karaman et al. (2011) evaluated the frequency of minor vs. major AE for 1305 lumbar TFESI performed in 562 patients utilizing intraoperative fluoroscopy; patients followed for up to 3 weeks demonstrated a 7.4% incidence of vascular penetration that resulted in a 11.5% incidence of minor AE (i.e., including vasovagal events), but no major AE.[ 15 ] Manchikanti (2015) et al. stated that most AE were attributed to cervical TFESI (2.4% of total ESI) largely attributed to; “...particulate steroid (emboli), arterial intimal flaps, arterial dissection, dislodgement of plaque causing embolism, arterial muscle spasm, and embolism of a fresh thrombus following disruption of the intima.”[ 16 ] Chang et al. (2020) found that most AE for lumbar TFESI were minor (2.4-9.6% incidence), and that major AE were only reported in individual case studies (i.e. including spinal abscesses, cord infarcts, or epidural hematomas).[ 5 ]

Variable Reporting of AE for Cervical and/or Lumbar ESI vs TFESI, with 2 of 3 Studies Emphasizing More AE with Cervical Injections

Three studies reported different frequencies for AE attributed to cervical and/or lumbar ESI vs. TFESI, with 2 citing greater AE for cervical injections (ESI/TFESI) [ Table 1 ].[ 11 , 16 , 17 ] McGrath et al. (2011) compared the incidence of AE in 3964 lumbar TFESI vs. 161 cervical ESI, 123 lumbar ESI, and 17 caudal injections, and identified fewer AE (2.1%) for cervical/lumbar TFESI (2.1%) vs cervical/lumbar ESI (6%).[ 17 ] Manchikanti (2015) et al. found more AE attributed to cervical TFESI (2.4% of total ESI).[ 16 ] In 2017, Epstein also observed more AE for cervical injections that included both CESI and C-TFESI vs. lumbar injections (i.e., CESI/C-TFESI including DT/CSF leaks, intramedullary cord injections, intravascular injuries/strokes, and others).[ 11 ]

Various Symptoms/Signs of DT/CSF Leaks and Other AE Following Preoperative Spinal Epidural Injections

Multiple symptoms and signs may signal DT/CSF leaks or other AE following preoperative cervical and/or lumbar epidural injections [ Table 1 ].[ 6 , 7 , 10 ] In 2013, Epstein quoted a 28% incidence of positional headaches, a 6-16% frequency of Adhesive Arachnoiditis, and a combined 7.9-11.6% risk for; intramedullary cord injections, intravascular injections/stroke, blindness, new neurological deficits/paralysis, clots/hematomas, seizures and death.[ 6 ] Also in 2013, Epstein discussed the varied etiologies of DT/CSF leaks encountered during spinal surgery, and/or documented with postoperative MR, CT, and Myelo-CT studies.[ 7 ] These included; preoperative epidural injections, intraoperative traumatic injuries due to lumbar resection of ossified yellow ligament, cervical removal of Ossification of the Posterior Longitudinal LIgament (OPLL), marsupialization of arachnoid cysts, revision surgeries, and postoperative scarring.[ 7 ] Out of a series of 336 patients undergoing average 4.7 level laminectomies/average 1.4 level non-instrumented fusions, Epstein (2015) found a total 7.14% (24 patients) incidence of DT/CSF fistulas; 7 were due to preoperative ESI. Other etiologies of these leaks included; 6 synovial cysts, 5 instances of ossification the yellow ligament (OYL), 3 cases of leaks due to postoperative scar, and 3 DT due to the resection of intradural tumors. Subtracting the 7 ESI and 3 patients with intradural tumors from the 24 overall incidence of DT/CSF leaks, reduced the frequency to 4.16% for the remaining 14 patients (6 synovial cysts, 5 OYL, 3 postoperative scar).

Frequency and Treatment/Repair of Traumatic Dural Tears (DT) Attributed to Preoperative Cervical/Lumbar Epidural Steroid Injections

Epstein documented in multiple studies that preoperative cervical and/or lumbar epidural injections resulted in intraoperatively documented DT that warranted direct repair [ Table 1 ].[ 6 - 8 , 10 , 13 ] In 2013, Epstein reviewed the literature, citing a 0.4-6.0% incidence of intraoperative CSF leaks following spinal epidural injections.[ 6 ] Also in 2013, Epstein discussed the varied etiologies of DT/CSF leaks documented with MR, CT, and Myelo-CT studies. Emphasis was placed on performing timely direct dural repairs utilizing 7-0 Gore-Tex interrupted sutures (i.e. suture larger than the needle occludes needle holes), supplemented by adjunctive techniques (i.e. muscle/fascial patch grafts, microfibrillar collagen (suturable/non-suturable microfibrillar collagen, fibrin sealants, lumbar drains, wound-peritoneal, and lumboperitoneal shunts)). [ 7 ] In a 2014 study, Epstein observed that 33 of 39 patients undergoing multilevel lumbar laminectomies and non-instrumented fusions had preoperative lumbar epidural injections (i.e. average of 4.1 ESI/patient performed 2-5 weeks preoperatively (avg. 3.9 weeks); 6 (18.2%) patients had confirmed intraoperative DT/CSF leaks warranting direct repairs.[ 8 ] Epstein (2015) later documented that for 336 patients undergoing average 4.7 level laminectomies/average 1.4 level non instrumented fusions (2000-2015), 7 (29.2%) of 24 intraoperative DT/CSF fistulas were due to preoperative epidural steroid injections.[ 10 ]

High Failure Rate for Epidural Blood Patches (EBP) Utilized to Occlude Lumbar DT/CSF Leaks Largely Attributed to Preoperative ESi/TFESI and Other Factors

In 2023, Epstein emphasized the need for early direct repair of MR/Myelo-CT-documented sites of DT/CSF leaks encountered intraoperatively following preoperative ESI and other procedures/factors (i.e. lumbar punctures, spinal anesthesia (SA), and spontaneous intracranial hypotension (SICH)) rather than choosing to perform EBP [ Table 1 ].[ 14 ] The efficacy of targeted EBP repair of DT/CSF leaks (i.e. typically utilizing intraoperative ultrasound, fluoroscopy, or the O-Arm) averaged 20%, with a range of from 0-60%.

Incidence of Acute Cervical Spinal Cord Injuries (SCI) Due to ESI/TFESI

Schreiber (2016) documented that 7 (0.52%) of 1343 patients admitted to an acute spinal cord injury (SCI) center sustained cervical injures attributed to ESI/TFESI (2001-2008) [ Table 1 ].[ 20 ] All 7 patients had incomplete acute cervical neurological deficits attributed to; 1 cord injection, 2 epidural abscesses, 1 cord contusion, 2 epidural hematomas, and 1 of unknown etiology. Acute presenting symptoms for these patients included; hypotension, respiratory decompensation, chest pain, upper extremity numbness, paresthesias, weakness, and/or fever.

Frequency of Epidural Hematomas Due to Epidural Injections/ESI

Out of a total of 1182 (0.51%) cervical and 4617 (0.15%) lumbar interlaminar ESI performed over an 8-year period, Smith et al. (2017) observed that 13 patients required emergency neurosurgery for 3 DT/CSF leaks, 3 infections, and 7 hematomas [ Table 1 ].[ 21 ] Interestingly, all 7 hematomas were lumbar cases; 3 patients had stopped anticoagulation, while 3 had continued on their Aspirin therapy.

Case Report: 3 Month Delay in L23 Laminectomy Due to Administration of 3 Lumbar ESI

In a 2015 case study, Epstein (2015) presented a 54-year-old-male with an MR-documented massive L23 disc herniation filling the spinal canal who was negligently treated for 3 months with ESI (i.e., 1 per month); when he finally presented paraplegic, he underwent a L23 laminectomy and fortunately recovered significant neurological function [ Table 1 ].[ 9 ]

CONCLUSION

In this perspective/review of the literature, patients undergoing epidural cervical and lumbar ESI or TFESI spinal injections demonstrated minimal to no short-term, and no long-term benefits for the treatment of cervical and/or lumbar pain/radiculopathy vs. placebos. Further, more AE were observed for cervical vs. lumbar epidural injections, with more frequent and severe AE seen with TFESI vs. ESI procedures.

Ethical approval

Institutional Review Board approval is not required.

Declaration of patient consent

Patient’s consent not required as there are no patients in this study.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

Use of artificial intelligence (AI)-assisted technology for manuscript preparation

The authors confirm that there was no use of artificial intelligence (AI)-assisted technology for assisting in the writing or editing of the manuscript and no images were manipulated using AI.

Commentary

Author: Jamie L. Baisden MD (Neurosurgery)

This article is a bit heavy on the risks of ESI and TFESI. What is this paper’s overall purpose? To let people know the dangers of epidural steroid injections lacking FDA approval, or to save insurance companies money? As spine surgeons, we should not allow insurance companies that don’t have a medical degree to act as gait-keepers to determine who should undergo multiple non-FDA-approved spinal epidural injections before being “disqualified” or “qualified” for spine surgery. Most oral steroids (i.e., oral steroids - Medrol Dose Packs, or Prednisone) or intramuscular steroid injections (i.e., especially trigger point injections) have minimal risks/minimal down-sides and may make people feel better, particularly in the short-term (i.e., weeks). Notably, the placebo effect of any injection (i.e., many studies typically compare epidural steroid vs Normal Saline epidural or intramuscular injections) is often around 30%; so, you may want to concede that 30% of patients may feel some transient improvement in the first 1-2 weeks. More critically, however, the “natural history” of spontaneous improvement kicks in at around 3-4 weeks, just around the time the “benefits” of steroids are actually waning or disappearing; patients may then mistakenly attribute their continued “improvement” to the steroids, rather than to the natural course of symptom resolution.

My primary concern, however, is what is left for patients if we can’t offer narcotics or epidural steroid injections anymore, and the patient can’t take NSAIDs (i.e., on blood thinners for cardiovascular disease, and/or a history of gastrointestinal, and/or chronic kidney disease)? We can certainly offer patient education and a multitude of medications (i.e., muscle relaxants, neuromodulators, non-opioid/non-NSAID medications) and/or other non-invasive modalities. It all remains a statistical balancing act of juggling potential risks versus benefits as we help patients navigate the cons/dangers posed by non-FDA-approved epidural spinal injections. Further, these unnecessary injections (i.e. well-documented minimal to no short-term (i.e. 3-6 weeks) and no long-term benefits) typically cost patients or their insurance carriers hundreds to over thousands of dollars per injection (i.e., varies by state, carrier, setting); certainly, these fees are lining pain management specialists’ pockets. Lastly, in some instances, epidural steroid injections are being wrongly performed, negligently delaying “essential” spine surgery.

Disclaimer

The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of the Journal or its management. The information contained in this article should not be considered to be medical advice; patients should consult their own physicians for advice as to their specific medical needs.

References

1. Abbasi A, Malhotra G, Malanga G, Elovic EP, Kahn S. Complications of interlaminar cervical epidural steroid injections: A review of the literature. Spine (Phila Pa 1976). 2007. 32: 2144-51

2. Anderberg L, Annertz M, Persson L, Brandt L, Saveland H. Transforaminal steroid injections for the treatment of cervical radiculopathy: A prospective and randomized study. Eur Spine J. 2007. 16: 321-8

3. Arden NK, Price C, Reading I, Stubbing J, Hazelgrove J, Dunne C. A multicenter randomized controlled trial of epidural corticosteroid injections for sciatica: the WEST study. Rheumatology (Oxford). 2005. 44: 1399-406

4. Carette S, Leclaire R, Marcoux S, Morin F, Blaise GA, St-Pierre A. Epidural corticosteroid injections for sciatica due to herniated nucleus pulposus. N Engl J Med. 1997. 336: 1634-40

5. Chang A, Ng AT. Complications associated with lumbar transforaminal epidural steroid injections. Curr Pain Headache Rep. 2020. 24: 67

6. Epstein NE. The risks of epidural and transforaminal steroid injections in the Spine: Commentary and a comprehensive review of the literature. Surg Neurol Int. 2013. 4: S74-93

7. Epstein NE. A review article on the diagnosis and treatment of cerebrospinal fluid fistulas and Dural tears occurring during spinal surgery. Surg Neurol Int. 2013. 4: S301-17

8. Epstein NE. Commentary: Unnecessary preoperative epidural steroid injections lead to cerebrospinal fluid leaks confirmed during spinal stenosis surgery. Surg Neurol Int. 2014. 5: S325-8

9. Epstein NE. Unnecessary multiple epidural steroid injections delay surgery for massive lumbar disc: Case discussion and review. Surg Neurol Int. 2015. 6: S383-7

10. Epstein NE. Incidence and management of cerebrospinal fluid fistulas in 336 multilevel laminectomies with noninstrumented fusions. Surg Neurol Int. 2015. 6: S463-8

11. Epstein NE. Neurological complications of lumbar and cervical Dural punctures with a focus on epidural injections. Surg Neurol Int. 2017. 8: 60

12. Epstein NE. Major risks and complications of cervical epidural steroid injections: An updated review. Surg Neurol Int. 2018. 9: 86

13. Epstein NE. Unnecessary cervical epidural injection in an octogenarian. Surg Neurol Int. 2019. 10: 108

14. Epstein NE, Agulnick MA. Perspective: Early direct repair of recurrent postoperative cerebrospinal (CSF) fluid leaks: No good evidence epidural blood patches (EBP) work. Surg Neurol Int. 2023. 14: 120

15. Karaman H, Kavak GO, Tufek A, Yldrm ZB. The complications of transforaminal lumbar epidural steroid injections. Spine (Phila Pa 1976). 2011. 36: E819-24

16. Manchikanti L, Hirsch JA. Neurological complications associated with epidural steroid injections. Curr Pain Headache Rep. 2015. 19: 482

17. McGrath JM, Schaefer MP, Malkamaki DM. Incidence and characteristics of complications from epidural steroid injections. Pain Med. 2011. 12: 726-31

18. Oliveira CB, Maher CG, Ferreira M, Hancock MJ, Oliveira VC, McLachlan A. Epidural corticosteroid injections for lumbosacral radicular pain. Cochrane Database Syst Rev. 2020. 4: CD013577

19. Parr AT, Diwan S, Abdi S. Lumbar interlaminar epidural injections in managing chronic low back and lower extremity pain: A systematic review. Pain Phys. 2009. 12: 163-88

20. Schreiber AL, McDonald BP, Kia F, Fried GW. Cervical epidural steroid injections and spinal cord injuries. Spine J. 2016. 16: 1163-6

21. Smith GA, Pace J, Strohl M, Kauo A, Hayek S, Miller JP. Rare neurosurgical complications of epidural injections: An 8-yr single-institution experience. Oper Neurosurg (Hagerstown). 2017. 13: 271-9

22. Valat JP, Giraudeau B, Rozenberg S, Goupille P, Bourgeois P, Micheau-Beaugendre V. Epidural corticosteroid injections for sciatica: A randomized, double blind, controlled clinical trial. Ann Rheum Dis. 2003. 62: 639-43