Review
Musculoskeletal Imaging
May 15, 2017

The Lumbar Neural Foramen and Transforaminal Epidural Steroid Injections: An Anatomic Review With Key Safety Considerations in Planning the Percutaneous Approach

Abstract

OBJECTIVE. The purpose of this article is to review the anatomy of the lumbar neural foramen and to describe techniques of transforaminal epidural steroid injections with emphasis on safety. Rare cases of paraplegia have been reported.
CONCLUSION. Although no consensus currently exists about which approach is the safest, knowledge of the foraminal anatomy is a key consideration when choosing a needle approach for transforaminal epidural steroid injections.
Epidural steroid injections are a cornerstone of conservative treatment of radiculopathy. These procedures have been performed since the 1950s and are the most frequently performed procedure in pain medicine in the United States [13]. Although they are rare, multiple cases have linked catastrophic spinal cord injuries to the use of particulate steroids administered via the transforaminal route [412]. These cases are thought to be due to unintentional intraarterial injection of steroid into a radiculomedullary artery that supplies the spinal cord, with resultant RBC agglutination and occlusion of the anterior spinal artery leading to cord infarction [7, 13]. Direct vascular trauma or vasospasm have also been suggested as factors possibly contributing to distal ischemic insult [14, 15]. This article will review the osseous and neurovascular anatomy of the neural foramen, with an emphasis on the radicular arteries, the artery of Adamkiewicz, and vascular variants. We will address the various approaches for performing transforaminal steroid injections within the context of the neural foraminal vascular anatomy.

Transforaminal Steroid Injections

The three main techniques for performing epidural steroid injections in the lumbar spine include transforaminal, interlaminar, and caudal approaches. The primary advantage of the transforaminal approach, the focus of this article, is the ability to deliver therapeutic agents as close as possible to the source of the pain. Meta-analyses have shown that transforaminal epidural steroid injections result in an improvement in pain [1618].
At least 18 cases of paralysis after transforaminal epidural steroid injections have been reported at every level of the lumbar spine as well as T12–S1, with both CT and fluoroscopic guidance [19]. Routine precautions to avoid intravascular injection including aspiration before injection of medications, visualization of normal epidural flow of contrast material, and use of digital subtraction angiography did not prevent spinal cord infarction in all cases [12]. Of the cases in which needle position could be determined, the needle was positioned most commonly in the superior portion of the neural foramen (77.7%) and less commonly in the midzone (22.2%); no cases were identified in the inferior portion of the neural foramen. Of these cases, the needle was most commonly anterior (71.4%) and less commonly posterior (28.5%).
Various techniques have been proposed to optimize safety with regard to avoidance of the nerve and vascular structures, including the posterolateral approach and Kambin triangle (infraneural) approach. Controversy is evolving regarding the safest approach for transforaminal epidural steroid injections to minimize the chance of paraplegia [19]. Regardless of the approach used, nonparticulate steroids (e.g., dexamethasone) are increasingly favored as the medication of choice for performing transforaminal steroid injections [15, 2022]. To our knowledge, no cases of paralysis have been reported with the use of non-particulate steroids. Preliminary results show dexamethasone to be safe, but long-term safety data of nonparticulate steroids for epidural use is lacking [23, 24]. The literature regarding the comparative efficacy of particulate and nonparticulate steroids continues to develop and most studies are small; however, a recent meta-analysis found no statistically significant difference in pain reduction or functional outcome between particulate and nonparticulate steroid formulations [25].
This review details the anatomy of the neural foramen, with particular focus on vascular anatomy and variants, to provide an anatomic rationale for various techniques of performing transforaminal epidural steroid injections.

Osseous Anatomy of the Lumbar Neural Foramen

The neural foramen (also called the intervertebral foramen) is an opening on either side of the spinal column at each intervertebral level through which the spinal nerve roots traverse while surrounded by arteries, veins, and epidural fat. The neural foramina are formed at the lateral aspects of the vertebral canal. The anterior border of the neural foramen is the intervertebral disk (along with the superior and inferior endplates of the adjacent vertebral bodies); the posterior border is the superior and inferior articular processes and facet joint; and the roof and floor of the neural foramen are formed by the pedicles of the respective levels (Fig. 1).
Fig. 1 —Sagittal illustration of two representative lumbar vertebral bodies and intervertebral disk shows osseous anatomy comprising neural foramen. Roof and floor of neural foramen are formed by pedicles of adjacent vertebrae. Anterior wall of neural foramen is intervertebral disk and adjacent endplates. Posterior wall is facet joint and inferior and superior articular processes.

Nerve Roots, Dorsal Root Ganglia, and Dura

The lumbar spinal nerves exit the spinal canal via the neural foramina and are numbered for the vertebra forming the roof of the neural foramen through which they exit. For instance, the L4 spinal nerve exits the neural foramen at the L4-5 disk level, with the pedicle of L4 forming the roof of the foramen. This numbering convention is in contrast to the cervical spine, where cervical nerves C1–C7 exit the neural foramina above the vertebral body for which they are named, whereas the C8 nerve and all thoracic and lumbar nerves exit below. Each nerve is formed by a dorsal and ventral root, which are covered by pia and a dural sleeve, with the nerve roots located in the intradural compartment (Fig. 2). A dorsal root ganglion is located just proximal to the junction of the dorsal and ventral roots, containing cell bodies of the sensory neural fibers in the dorsal root. The ventral and dorsal roots are connected with a plexus of fascicles [26]. The nerve roots join to form a spinal nerve distal and lateral to the dorsal root ganglion, which is covered by perineurium. Each spinal nerve divides into a larger ventral ramus and smaller dorsal ramus just outside the foramen. Knowledge of this dural anatomy is an important consideration when performing transforaminal epidural steroid injections because the risk of an inadvertent dural puncture increases if the medial portion of the foramen is encountered.
Fig. 2 —Transverse illustration of representative upper lumbar foraminal region shows ventral and dorsal nerve roots arising from ventral and dorsal surface of spinal cord, respectively. Nerve roots and dorsal root ganglia are intradural and are located within medial portion of neural foramen.

Radicular and Radiculomedullary Arteries and Arterial Supply to the Spinal Cord

The blood supply of each of the 31 bilateral nerve roots (eight cervical, 12 thoracic, five lumbar, five sacral, and one sacrococcygeal root per side) is supplied in a segmental manner by a radicular artery accompanying each nerve root. In the thoracic and lumbar spine, these radicular arteries are derived from branches of the aorta: intercostal arteries in the thoracic spine and lumbar arteries in the lumbar spine. The intercostal and lumbar arteries divide into a dorsal segment supplying the paraspinal muscles, a somatic branch ventral to the spinal canal that feeds the dura, and a radicular artery that supplies the spinal nerve and nerve roots within the neural foramina (Fig. 3). A radicular artery typically divides into an anterior and posterior radicular artery, but this division and the location of the radicular artery or arteries within the neural foramen vary. The caliber of the radicular arteries varies between 0.2 mm and 2 mm, in comparison with the outer diameter of a typical 22-gauge spinal needle used commonly for epidural steroid injections, which is approximately 0.72 mm [27]. A recent retrospective review evaluated the location of radicular arteries in the neural foramina of 32 patients who had undergone abdominal CT angiography for aortic disease [28]. This study found more than twice the incidence of radicular arteries in the superior portion of the foramen (50.4%) compared with the inferior portion (20.3%). However, this study did not state the criteria for distinguishing between an artery and a vein, which can be a difficult distinction [29].
Fig. 3 —Transverse illustration of upper lumbar vertebral body shows arterial supply of radicular and radiculomedullary arteries arising from lumbar arteries. Note that although radicular artery is present at every level, laterality and level of radiculomedullary arteries, which continue to supply spinal cord, are highly variable. On this illustration, anterior and posterior radiculomedullary arteries are drawn on left. Largest and most caudal anterior radiculomedullary artery is artery of Adamkiewicz.
Of note, an end artery that supplies the neural elements within the neural foramen is a radicular artery; if that artery continues to supply the spinal cord, it is a termed a radiculomedullary artery. Although radicular arteries at every spinal level supply the foraminal structures including nerves, dura, and vertebral bodies, most are end arteries that do not contribute to the arterial supply of the spinal cord. Relatively few radiculo medullary arteries supply the spinal cord, because most regress during fetal development.
The arterial supply of the spinal cord can be divided into anterior and posterior axes. The anterior axis is composed of the anterior spinal artery, which courses along the ventral midline of the spinal cord and feeds the anterior two-thirds of the spinal cord. The paired posterolateral axes provide arterial supply to the posterior third of the cord [30]. In the entire spinal cord, approximately four to eight anterior radiculomedullary arteries supply the anterior spinal artery, and 10–20 posterior radiculomedullary arteries supply the paired posterior spinal arteries [30].

Artery of Adamkiewicz

The anterior spinal artery is supplied by the anterior radiculomedullary arteries, the largest and usually the most caudal of which is the artery of Adamkiewicz, also known as the arteria radicularis magna. In adults, the vascularization of the lower spinal cord caudad to the thoracolumbar junction is highly dependent on the artery of Adamkiewicz, because the anterior spinal artery usually narrows cranial to the anastomosis with the artery of Adamkiewicz [31]. The origin of the artery of Adamkiewicz is highly variable but most commonly originates on the left (68–85%). The artery of Adamkiewicz usually arises between T9 and L5 [32], most commonly at T9 and less commonly below L2 (23.5%) [33]. The diameter of the artery of Adamkiewicz ranges from 0.6 to 1.2 mm [33]. In cases in which the artery of Adamkiewicz has a high thoracic origin, enlargement of an iliac-derived radiculomedullary artery may contribute to distal spinal cord blood supply, entering a neural foramen as caudad as S1 [5].
Similar to the radicular arteries, the artery of Adamkiewicz is most commonly located in the ventral superior or midportion of the neural foramen. In a total of 34 cadavers in two separate studies, the artery of Adamkiewicz was located in the lower third of the foramen in only a single instance [34, 35]. The artery was located in the superior half of the foramen in 97% of cases in a separate angiographic study [36].
Once the artery of Adamkiewicz enters the intradural space, it courses superiorly and makes a characteristic hairpin turn before anastomosing with the anterior spinal artery [37] (Fig. 4). The artery of Adamkiewicz can be visualized with MR and CT angiography, on which it is identified as continuous with the aorta via an intercostal or lumbar artery and radiculomedullary artery and ascending to the anterior midsagittal surface of the spinal cord to anastomose with the anterior spinal artery [29, 3840]. However, the artery of Adamkiewicz can be difficult to visualize on noninvasive vascular imaging, often requiring postprocessing to be reliably seen [39]. To our knowledge, no data or consensus recommendation supports vascular imaging (e.g., CT or MR angiography) before transforaminal steroid injection.
Fig. 4 —Illustration of anterior aspect of spinal cord shows artery of Adamkiewicz, which in this case arises from left lumbar radiculomedullary artery. Note radicular arteries drawn at other levels, which do not contribute to spinal cord blood supply. Anterior spinal artery (ASA) variably narrows cranial to anastomosis of artery of Adamkiewicz and anterior spinal artery.

Spinal Cord and Foraminal Veins

Unlike the arterial system, the anterior and posterior venous systems of the spinal cord are codominant. In addition to the anterior and posterior veins, small lateral veins also run dorsally [30]. These dorsal, ventral, and lateral veins empty into anterior and posterior radicular veins running along the nerve roots and anterior and posterior epidural plexuses (Fig. 5). Systemic venous drainage is through lumbar veins, which drain into the inferior vena cava, and the ascending lumbar vein on the left, which drains into the left renal vein. The basivertebral veins connect the epidural venous plexus and drainage from the posterior aspect of the vertebral body to the lumbar vein.
Fig. 5 —Transverse illustration of upper lumbar vertebral body shows venous anatomy of spinal cord and foraminal region. Spinal cord is drained by anterior-ventral and posterior-dorsal veins and paired lateral veins, which form rich anterior and posterior epidural plexuses. Communication with lumbar veins is via anterior and posterior foraminal veins and median basivertebral vein. Lumbar veins drain into inferior vena cava (IVC) and left ascending lumbar vein drains into left renal vein.

Choice of Modality to Perform Transforaminal Epidural Steroid Injection

Transforaminal epidural steroid injections may be performed with fluoroscopic or CT guidance [41, 42]; however, to our knowledge no studies have compared the relative effectiveness or safety of these two modalities, and paralysis has occurred with both CT and fluoroscopic guidance. Generally, the choice of modality reflects resource availability, training experience, and institutional factors [43]. Advantages of fluoroscopy include its wide availability, the ability to visualize the pattern of contrast material flow in real time, and the utility of a C-arm to allow needle paths in nonaxial planes, a typical limitation of CT. The main advantage of CT is excellent 3D imaging of soft tissue and osseous anatomy, with precise visualization of needle placement [44]. An additional advantage is absence of operator radiation exposure if CT fluoroscopy is not used. The primary disadvantages of CT in comparison with fluoroscopy include increased radiation dose to the patient, longer procedure time, increased cost, and most importantly the inability to visualize vascular flow, although CT fluoroscopy has been proposed as means to image real-time contrast material flow [45].

Safe Triangle Approach

The traditional needle target for transforaminal injection is the epidural space just caudad to the inferior margin of the pedicle, immediately superior, lateral, and anterior to the targeted exiting nerve. On the lateral oblique fluoroscopic images, the target area forms a triangle bordered by the inferior margin of the pedicle, the exiting nerve root (forming the hypotenuse of the triangle), and a line drawn inferiorly from the anterior margin of the pedicle (Fig. 6). This approach, usually referred to as the safe triangle approach but also termed the subpedicular or supraneural approach, was originally described as an approach where injection could be performed with minimal risk of nerve injury, intrathecal puncture, or vascular injection [46]. The needle crosses anterior to the nerve and is advanced to the dorsal periosteum of the vertebral body, where medication is administered in the anterior epidural space [47, 48].
Fig. 6 —Oblique illustration of two lumbar vertebral bodies showing traditional safe triangle, colored in blue. Roof of safe triangle is inferior margin of pedicle, hypotenuse is exiting nerve root, and anterior border of safe triangle is created by line drawn inferiorly from anterior margin of pedicle.
On the fluoroscopic anteroposterior view, the needle should remain lateral to the midpedicular line; positioning the needle more medially in the neural foramen increases the risk for dural puncture [46]. If the needle is positioned too laterally, a selective nerve root block may be performed, and retrograde epidural flow may not be achieved. After epidural contrast material is administered, it should outline the nerve root sheath and retrograde or medial epidural flow (Fig. 7).
Fig. 7A —47-year-old man with right lumbar radiculopathy undergoing right L5–S1 transforaminal fluoroscopic-guided epidural steroid injection via traditional safe triangle approach.
A, Lateral oblique needle trajectory view shows needle tip (arrow) in safe triangle inferior to pedicle.
Fig. 7B —47-year-old man with right lumbar radiculopathy undergoing right L5–S1 transforaminal fluoroscopic-guided epidural steroid injection via traditional safe triangle approach.
B, Lateral fluoroscopic image shows needle tip (arrow) advanced to dorsal periosteum of vertebral body.
Fig. 7C —47-year-old man with right lumbar radiculopathy undergoing right L5–S1 transforaminal fluoroscopic-guided epidural steroid injection via traditional safe triangle approach.
C, Anteroposterior view after injection of 2 mL of Omnipaque 180 (GE Healthcare) shows epidural (single asterisk) and perineural (double asterisks) flow of contrast material. Needle tip remains lateral to midpedicular line. Patient experienced pain relief after procedure and no complications were noted.
If CT guidance is used, the target is the anterior epidural space within the anterior margin of the neural foramen, nearly abutting the dorsal periosteum of the vertebral body just inferior to the pedicle. After injection of contrast material, medial epidural and perineural flow of contrast material should be seen (Fig. 8).
Fig. 8A —57-year-old woman with left lower extremity pain undergoing left L2–L3 transforaminal epidural steroid injection with traditional safe triangle approach under CT guidance.
A, Scout CT topogram with grid placed on overlying skin is annotated with final needle tip position (caliper) determined on PACS workstation after procedure, with needle tip just inferior to pedicle of L2. Posterior instrumented fusion hardware is present at L3–L4.
Fig. 8B —57-year-old woman with left lower extremity pain undergoing left L2–L3 transforaminal epidural steroid injection with traditional safe triangle approach under CT guidance.
B, Axial intraprocedural CT image shows final needle tip position (arrow), which is in anterior epidural space abutting dorsal periosteum of vertebral body. Contrast material (0.3 mL Omnipaque 180, GE Healthcare) was injected, showing central epidural (single asterisk) and perineural flow (double asterisks). Patient experienced pain relief after procedure and no complications were noted.
Recent literature has questioned the safety of the safe triangle approach because of the location of the radicular or radiculomedullary artery in the anterosuperior portion of the foramen and the reported cases of paraplegia associated with this technique [19, 35, 36].

Posterolateral Approach

The posterolateral approach is a modification of the safe triangle approach with the needle tip remaining in the posterior portion of the neural foramen (Fig. 9) and slightly inferior compared with the safe triangle approach [46, 49]. On the lateral oblique fluoroscopic trajectory view, the target remains within the safe triangle. However, the needle is not advanced into the anterior epidural space; rather, the tip stays within the posterior aspect of the neural foramen (Fig. 10), as confirmed on lateral fluoroscopic views. An injection planned with the traditional safe triangle approach can easily be modified to the posterolateral approach for cases of severe foraminal stenosis or if the patient has nerve pain during the procedure and the anterior epidural space cannot be accessed.
Fig. 9 —Lateral illustration of two lumbar vertebral bodies showing needle position of posterolateral approach (blue needle), in comparison with needle position of traditional safe triangle approach (gray needle). Note relative location of exiting nerve root and radicular artery located anterosuperior to nerve.
Fig. 10A —39-year-old man with right lumbar radiculopathy undergoing right L5–S1 transforaminal epidural steroid injection with posterolateral approach under fluoroscopic guidance.
A, Lateral oblique needle trajectory view shows needle laid over skin (arrow) confirming needle entry site, slightly more inferior than traditional safe triangle approach.
Fig. 10B —39-year-old man with right lumbar radiculopathy undergoing right L5–S1 transforaminal epidural steroid injection with posterolateral approach under fluoroscopic guidance.
B, Lateral fluoroscopic image shows final needle tip position (arrow), remaining within posterior half of neural foramen. Tiny test dose of contrast material was administered (asterisk).
Fig. 10C —39-year-old man with right lumbar radiculopathy undergoing right L5–S1 transforaminal epidural steroid injection with posterolateral approach under fluoroscopic guidance.
C, Anteroposterior view after injection of 2 mL of Omnipaque 180 (GE Healthcare) shows epidural (single asterisk) and perineural (double asterisks) flow of contrast material. Needle tip (arrow) remains lateral to midpedicular line. No complications were noted.
For CT-guided posterolateral injections, the target is the posterior epidural fat in the posterior and lateral aspect of the neural foramen. Under CT, the exiting nerve can often be visualized and the needle positioned at its intended target just posterior to the nerve, which may be an advantage over fluoroscopy, during which the exact position of the nerve is not known. As with the safe triangle approach, epidural and perineural flow of contrast material should be seen (Fig. 11).
Fig. 11A —70-year-old woman with chronic back pain and superimposed right L5–S1 radicular component undergoing CT-guided right L5–S1 transforaminal epidural steroid injection with posterolateral approach under CT guidance. Posterolateral approach was chosen due to foraminal stenosis.
A, Scout CT topogram with grid placed on overlying skin is annotated with final needle tip position (caliper) determined on PACS workstation after procedure, with needle tip in posterior aspect of foramen at L5–S1, slightly inferior to L5 pedicle.
Fig. 11B —70-year-old woman with chronic back pain and superimposed right L5–S1 radicular component undergoing CT-guided right L5–S1 transforaminal epidural steroid injection with posterolateral approach under CT guidance. Posterolateral approach was chosen due to foraminal stenosis.
B, Axial intraprocedural CT image shows tip of spinal needle located within posterior margin of neural foramen (arrow). Epidural (single asterisk) and perineural (double asterisks) flow of contrast material is seen after injection of 0.3 mL Omnipaque 180 (GE Healthcare). No complications were noted.
A single-center prospective study of 50 patients showed no significant differences in outcome measures between the safe triangle and posterolateral approaches [50], but less nerve pain was reported with the posterolateral approach. The posterolateral approach has a theoretically decreased risk of arterial injury because it avoids the anterosuperior portion of the foramen; however, further studies are needed to show that the posterolateral approach is safer than the traditional safe triangle approach in this regard.

Kambin Triangle Approach

The Kambin triangle is a triangular space located over the dorsal aspect of the intervertebral disk. It was first described in 1986 as a safe portal for percutaneous lateral diskectomy [51]. The triangle is bounded inferiorly by the superior endplate of the inferior vertebral body; the hypotenuse of the triangle is formed by the exiting nerve root; and the posterior margin of the triangle is a line formed by the endplate inferiorly and the superior articulating facet superiorly (Fig. 12). This space has been proposed as a safe and convenient access for percutaneous diskectomy, diskograms, and safe access for transforaminal steroid injections [52, 53].
Fig. 12 —Oblique illustration of two lumbar vertebral bodies showing Kambin triangle, colored in green. Kambin triangle is bounded inferiorly by superior endplate of inferior vertebral body; hypotenuse of triangle is formed by exiting nerve root; and posterior margin of triangle is line formed by endplate inferiorly and superior articulating facet superiorly.
On the lateral oblique fluoroscopic trajectory images, the target should be the inferior one-third of the neural foramen, at the level of the intervertebral disk [53]. The initial trajectory for this approach is the same as for diskography, such that if the needle is advanced too far ventrally during epidural injection, an inadvertent intradiskal injection may result [54]. Similar to the safe triangle approach, on the frontal radiograph the needle should not be advanced medial to the midpedicular line to avoid intradural puncture. After the injection of contrast material, perineural and epidural flow should be seen (Fig. 13).
Fig. 13A —58-year-old woman with back pain undergoing fluoroscopically guided right L5–S1 transforaminal epidural steroid injection with Kambin triangle approach under fluoroscopic guidance.
A, Lateral oblique trajectory fluoroscopic image shows needle tip (arrow) in inferior portion of foramen at level of intervertebral disk.
Fig. 13B —58-year-old woman with back pain undergoing fluoroscopically guided right L5–S1 transforaminal epidural steroid injection with Kambin triangle approach under fluoroscopic guidance.
B, Lateral fluoroscopic image shows needle tip (arrow) at level of L5-S1 disk but dorsal to posterior spinal line.
Fig. 13C —58-year-old woman with back pain undergoing fluoroscopically guided right L5–S1 transforaminal epidural steroid injection with Kambin triangle approach under fluoroscopic guidance.
C, After injection of 2 mL of Omnipaque 180 (GE Healthcare), anteroposterior fluoroscopic image shows medial epidural (single asterisk) and perineural (double asterisks) flow of contrast material. No complications were noted.
For CT-guided Kambin triangle injections, the target is the anterior epidural space just dorsal to the intervertebral disk. Correlation with the topographic CT, acquired in the lateral orientation, is especially helpful to confirm that the needle entry site is in plane with the level of the intervertebral disk. Similar to fluoroscopically guided procedures, to prevent inadvertent disk injection, care must be taken not to advance the needle too ventrally. Like the other needle approaches, injected contrast material should show epidural and perineural flow (Fig. 14).
Fig. 14A —75-year-old woman with right-sided low back pain radiating down right leg undergoing right L4–L5 transforaminal epidural steroid injection with Kambin triangle approach under CT guidance.
A, Scout CT topogram with grid placed on overlying skin is annotated with final needle tip position (caliper) determined on PACS workstation after procedure, with needle tip at level of L4-5 intervertebral disk.
Fig. 14B —75-year-old woman with right-sided low back pain radiating down right leg undergoing right L4–L5 transforaminal epidural steroid injection with Kambin triangle approach under CT guidance.
B, Axial intraprocedural image shows final needle tip position, which is in anterior epidural space at level of intervertebral disk (arrows). After injection of 0.4 mL Omnipaque 180 (GE Healthcare), central epidural flow in anterior epidural space (single asterisk) and peripheral perineural flow (double asterisks) are seen. No complications were noted.
A single-institution retrospective review of 257 transforaminal epidural steroid injections using a Kambin triangle approach found a 4.7% intradiskal injection rate, 3.1% intrathecal injection rate, and 6.6% vascular injection rate (without distinguishing between arteries and veins) [55] compared with the safe triangle approach, in which the rate of disk injection has been reported to be lower, ranging between 0.17% and 2.3% [56, 57]. Intrathecal injections with the safe triangle approach are rare, with a reported incidence of 0.04% [58]. The incidence of intravascular injections with the safe triangle approach is 11.2% [59].
A single-institution prospective study of 100 consecutive cases found no differences in functional improvement between the traditional safe triangle approach and the Kambin triangle approach [60]. Vascular injection and intervertebral disk injections occurred with both techniques (vascular injection occurred in 4% for Kambin triangle and 12% for safe triangle; disk injection occurred in 2% for Kambin triangle and 4% for safe triangle), although statistical analysis on these adverse events was not performed.
A variant of the Kambin triangle approach is the preganglionic approach, which targets the supraadjacent level with an inferior foraminal approach. This technique can be used for cases in which there is nerve root compression by a disk at the superior level. For instance, if the L5 nerve root is compressed by a disk bulge at the L4-5 disk level, an L4–L5 preganglionic or inferior transforaminal epidural steroid injection can be performed. In this setting, the preganglionic approach has been shown to have increased efficacy, but large-scale safety studies are needed [61, 62].

Conclusion

Knowledge of the osseous and neurovascular anatomy of the neural foramen and blood supply to the spine is critical when undertaking spinal interventions, especially transforaminal steroid injections. Although these procedures are generally safe, rare reports of paraplegia after transforaminal steroid injection warrant caution.
Particulate steroid injections were implicated in all reported cases of paraplegia [19]. Of the cases in which needle position could be determined, the needle was positioned most commonly in the superior portion of the foramen. To our knowledge, no cases of paralysis have been reported with the needle in the inferior portion of the foramen or with the use of nonparticulate steroids (e.g., dexamethasone), which are increasingly favored. However, additional studies are needed to evaluate whether these procedural changes reduce the incidence of paraplegia, because most lumbar transforaminal steroid injections have been performed with the safe triangle approach, the use of nonparticulate steroids is a relatively recent trend, and direct vascular injury or vasospasm may also be a factor in these devastating injuries.
No single specific technique has been demonstrated as the safest for performing transforaminal epidural steroid injections in all patients. Additionally, the needle path is often dictated by patient anatomy, patient tolerance of needle positioning, or a combination of those factors, which cannot be reliably predicted in planning the procedure. However, several authors have recommended caution regarding the traditional safe triangle approach, especially if an alternative approach is feasible [19, 35, 36]. By reviewing the neurovascular anatomy of the neural foramen and the multiple available techniques for performing transforaminal epidural steroid injections, spine interventionalists can better determine the safest and most effective technique on the basis of the specific anatomic considerations of each patient.

Footnote

WEB
This is a web exclusive article.

References

1.
Hession WG, Stanczak JD, Davis KW, Choi JJ. Epidural steroid injections. Semin Roentgenol 2004; 39:7–23
2.
Mathis JM. Epidural steroid injections. Neuroimaging Clin N Am 2010; 20:193–202
3.
Bicket MC, Gupta A, Brown C, Cohen SP. Epidural injections for spinal pain. Anesthesiology 2013; 119:907–931
4.
Houten JK, Errico TJ. Paraplegia after lumbosacral nerve root block: report of three cases. Spine J 2002; 2:70–75
5.
Huntoon MA, Martin DP. Paralysis after transforaminal epidural injection and previous spinal surgery. Reg Anesth Pain Med 2004; 29:494–495
6.
Somayaji HS, Saifuddin A, Casey ATH, Briggs TWR, Orth F. Spinal cord infarction following therapeutic computed tomography-guided left L2 nerve root injection. Spine (Phila Pa 1976) 2005; 30:E106–E108
7.
Neal JM. Anatomy and pathophysiology of spinal cord injury associated with regional anesthesia and pain medicine. Reg Anesth Pain Med 2008; 33:423–434
8.
Kennedy DJ, Dreyfuss P, Aprill CN, Bogduk N. Paraplegia following image-guided transforaminal lumbar spine epidural steroid injection: two case reports. Pain Med 2009; 10:1389–1394
9.
Glaser SE, Shah RV. Root cause analysis of paraplegia following transforaminal epidural steroid injections: the “unsafe” triangle. Pain Physician 2010; 13:237–244
10.
Helm S, Glaser S, Falco F, Henry B. A medical-legal review regarding the standard of care for epidural injections, with particular reference to a closed case. Pain Physician 2010; 13:145–150
11.
Thefenne L, Dubecq C, Zing E, et al. A rare case of paraplegia complicating a lumbar epidural infiltration. Ann Phys Rehabil Med 2010; 53:575–583
12.
Chang Chien GC, Candido KD, Knezevic NN. Digital subtraction angiography does not reliably prevent paraplegia associated with lumbar transforaminal epidural steroid injection. Pain Physician 2012; 15:515–523
13.
Laemmel E, Segal N, Mirshahi M, et al. Deleterious effects of intra-arterial administration of particulate steroids on microvascular perfusion in a mouse model 1. Radiology 2016; 279:731–740
14.
Ranson MT, Deer TR. Epidural injections for the treatment of spine-related pain syndromes. In: Mathis JM, Golovac S, eds. Image-guided spine interventions, 2nd ed. New York, NY: Springer, 2010:157–174
15.
Rathmell JP, Benzon HT, Dreyfuss P, et al. Safeguards to prevent neurologic complications after epidural steroid injections. Anesthesiology 2015; 122:974–984
16.
MacVicar J, King W, Landers MH, Bogduk N. The effectiveness of lumbar transforaminal injection of steroids: a comprehensive review with systematic analysis of the published data. Pain Med 2013; 14:14–28
17.
Quraishi NA. Transforaminal injection of corticosteroids for lumbar radiculopathy: systematic review and meta-analysis. Eur Spine J 2012; 21:214–219
18.
Buenaventura RM, Datta S, Abdi S, Smith HS. Systematic review of therapeutic lumbar transforaminal epidural steroid injections. Pain Physician 2009; 12:233–251
19.
Atluri S, Glaser SE, Shah RV, Sudarshan G. Needle position analysis in cases of paralysis from transforaminal epidurals: consider alternative approaches to traditional technique. Pain Physician 2013; 16:321–334
20.
Diehn FE, Murthy NS, Maus TP. Science to practice: what causes arterial infarction in transforaminal epidural steroid injections, and which steroid is safest? Radiology 2016; 279:657–659
21.
MacMahon PJ, Huang AJ, Palmer WE. Spine injectables: what is the safest cocktail? AJR 2016; 207:526–533
22.
Dietrich TJ, Sutter R, Froehlich JM, Pfirrmann CWA. Particulate versus non-particulate steroids for lumbar transforaminal or interlaminar epidural steroid injections: an update. Skeletal Radiol 2015; 44:149–155
23.
El Abd O, Amadera J, Pimentel D, Gomba L. Immediate and acute adverse effects following transforaminal epidural steroid injections with dexamethasone. Pain Physician 2015; 18:277–286
24.
Benzon HT, Chew TL, McCarthy RJ, Benzon HA, Walega DR. Comparison of the particle sizes of different steroids and the effect of dilution. Anesthesiology 2007; 106:331–338
25.
Mehta P, Syrop I, Singh JR, Kirschner J. Systematic review of the efficacy of particulate vs non particulate corticosteroids in epidural injections. PM&R 2016 Nov 30 [Epub ahead of print]
26.
Kostelic JK, Haughton VM, Sether L. Lumbar spinal nerves in the neural foramen: MR appearance. Radiology 1991; 178:837–839
27.
Demondion X, Lefebvre G, Fisch O, Vandenbussche L, Cepparo J, Balbi V. Radiographic anatomy of the intervertebral cervical and lumbar foramina (vessels and variants). Diagn Interv Imaging 2012; 93:690–697
28.
Simon JI, Mcauliffe M, Smoger D. Location of radicular spinal arteries in the lumbar spine from analysis of CT angiograms of the abdomen and pelvis. Pain Med 2016; 17:46–51
29.
Melissano G, Chiesa R. Advances in imaging of the spinal cord vascular supply and its relationship with paraplegia after aortic interventions. a review. Eur J Vasc Endovasc Surg 2009; 38:567–577
30.
Wakhloo AK, Patel NV, DeLeo MJ III, Shaibani A. Vascular anatomy of the spine, imaging, and endovascular treatment of spinal vascular diseases. In: Herkowitz HN, Garfin SR, Eismont FJ, Bell GR, Balderston RA. Rothman-Simeone the spine, 6th ed., vol. 2. Philadelphia, PA: Saunders, 2011:1655–1687
31.
Morishita K, Murakami G, Fujisawa Y, et al. Anatomical study of blood supply to the spinal cord. Ann Thorac Surg 2003; 76:1967–1971
32.
Biglioli P, Spirito R, Roberto M, et al. The anterior spinal artery: the main arterial supply of the human spinal cord—a preliminary anatomic study. J Thorac Cardiovasc Surg 2000; 119:376–379
33.
Rodriguez-Baeza A, Muset-Lara A, Rodriguez-Pazos M, Domenech-Mateu JM. The arterial supply of the human spinal cord: a new approach to the arteria radicularis magna of Adamkiewicz. Acta Neurochir (Wien) 1991; 109:57–62
34.
Alleyne CH, Cawley CM, Shengelaia GG, Barrow DL. Microsurgical anatomy of the artery of Adamkiewicz and its segmental artery. J Neurosurg 1998; 89:791–795
35.
Kroszczynski AC, Kohan K, Kurowski M, Olson TR, Downie SA. Intraforaminal location of thoracolumbar anterior medullary arteries. Pain Med 2013; 14:808–812
36.
Murthy NS, Maus TP, Behrns CL. Intraforaminal location of the great anterior radiculomedullary artery (artery of Adamkiewicz): a retrospective review. Pain Med 2010; 11:1756–1764
37.
Uflacker R. Atlas of vascular anatomy: an angiographic approach. Baltimore, MD: Williams & Wilkins, 1997:119–124
38.
Nijenhuis RJ, Mull M, Wilmink JT, Thron AK, Backes WH. MR angiography of the great anterior radiculomedullary artery (Adamkiewicz artery) validated by digital subtraction angiography. AJNR 2006; 27:1565–1572
39.
Melissano G, Bertoglio L, Civelli V, et al. Demonstration of the Adamkiewicz artery by multidetector computed tomography angiography analysed with the open-source software OsiriX. Eur J Vasc Endovasc Surg 2009; 37:395–400
40.
Yoshioka K, Niinuma H, Ohira A, et al. MR angiography and CT angiography of the artery of Adamkiewicz: noninvasive preoperative assessment of thoracoabdominal aortic aneurysm. RadioGraphics 2003; 23:1215–1225
41.
Silbergleit R, Mehta BA, Sanders WP, Talati SJ. Imaging-guided injection techniques with fluoroscopy and CT for spinal pain management. RadioGraphics 2001; 21:927–939
42.
Aguirre DA, Bermudez S, Diaz OM. Spinal CT-guided interventional procedures for management of chronic back pain. J Vasc Interv Radiol 2005; 16:689–697
43.
Palmer WE. Spinal injections for pain management. Radiology 2016; 281:669–688
44.
Chang A, Pochert S, Romano C, Brook A, Miller T. Safety of 1000 CT-guided steroid injections with air used to localize the epidural space. AJNR 2011; 32:175–177
45.
Wagner AL. CT fluoroscopy–guided epidural injections: technique and results. AJNR 2004; 25:1821–1823
46.
Berkwits L, Davidoff SJ, Buttaci CJ, Furman MB. Lumbar transforaminal epidural steroid (traditional) approach. In: Furman MB, Lee TS, Berkwits L, eds. Atlas of image-guided spinal procedures. Philadelphia, PA: Saunders, 2012:93–103
47.
Bogduk N. Lumbar transforaminal injection of corticosteroids. In: Practice guidelines for spinal diagnostic and treatment procedures. San Francisco, CA: International Spine Intervention Society, 2004:171–173
48.
Huston CW, Slipman CW, Furman MB, Hasan S, Derby R. Spinal injections. In: Slipman CW, Derby R, Simeone FA, Mayer TG, eds. Interventional spine: an algorithmic approach, 1st ed. Philadelphia, PA, Saunders, 2008:245–273
49.
Lee IS, Kim SH, Lee JW, et al. Comparison of the temporary diagnostic relief of transforaminal epidural injection approaches: conventional versus posterolateral technique. AJNR 2007; 28:204–208
50.
Park JW, Nam HS, Park Y. Usefulness of posterolateral transforaminal approach in lumbar radicular pain. Ann Rehabil Med 2011; 35:395–404
51.
Kambin P, Sampson S. Posterolateral percutaneous suction-excision of herniated lumbar intervertebral discs: report of interim results. Clin Orthop Relat Res 1986; 207:37–43
52.
Alamin T, Malek F, Carragee E, Kim MJ. The functional anaesthetic discogram: description of a novel diagnostic technique and report of 3 cases. SAS J 2008; 2:107–113
53.
Petrolla JJ, Furman MB. Lumbar transforaminal epidural steroid injection, infraneural approach. In: Furman MB, Lee TS, Berkwits L, eds. Atlas of image-guided spinal procedures. Philadelphia, PA: Saunders, 2012:105–110
54.
Trinh KH, Gharibo CG, Aydin SM. Inadvertent intradiscal injection with TFESI utilizing Kambin's retrodiscal approach in the treatment of acute lumbar radiculopathy. Pain Pract 2016; 16:E70–E73
55.
Levi D, Horn S, Corcoran S. The incidence of intradiscal, intrathecal, and intravascular flow during the performance of retrodiscal (infraneural) approach for lumbar transforaminal epidural steroid injections. Pain Med 2016; 17:1416–1422
56.
Plastaras CT, Casey E, Goodman BS, Chou L, Roth D, Rittenberg J. Inadvertent intradiscal contrast flow during lumbar transforaminal epidural steroid injections: a case series examining the prevalence of intradiscal injection as well as potential associated factors and adverse events. Pain Med 2010; 11:1765–1773
57.
Hong JH, Kim SY, Huh B, Shin HH. Analysis of inadvertent intradiscal and intravascular injection during lumbar transforaminal epidural steroid injections: a prospective study. Reg Anesth Pain Med 2013; 38:520–525
58.
El-Yahchouchi CA, Plastaras CT, Maus TP, et al. Adverse event rates associated with transforaminal and interlaminar epidural steroid injections: a multi-institutional study. Pain Med 2016; 17:239–249
59.
Furman MB, O'Brien EM, Zgleszewski TM. Incidence of intravascular penetration in transforaminal lumbosacral epidural steroid injections. Spine (Phila Pa 1976) 2000; 25:2628–2632
60.
Park KD, Lee J, Jee H, Park Y. Kambin triangle versus the supraneural approach for the treatment of lumbar radicular pain. Am J Phys Med Rehabil 2012; 91:1039–1050
61.
Lee JW, Sung HK, Choi JY. Transforaminal epidural steroid injection for lumbosacral radiculopathy: preganglionic versus conventional approach. Korean J Radiol 2006; 7:139–144
62.
Jeong HS, Lee JW, Kim SH, Myung JS, Kim JH, Kang HS. Effectiveness of transforaminal epidural steroid injection by using a preganglionic approach: a prospective randomized controlled study. Radiology 2007; 245:584–590

Information & Authors

Information

Published In

American Journal of Roentgenology
Pages: W26 - W35
PubMed: 28504548

History

Submitted: September 26, 2016
Accepted: January 17, 2017
First published: May 15, 2017

Keywords

  1. epidural steroid injection
  2. lumbar neural foramen
  3. radiculopathy
  4. spine

Authors

Affiliations

Jacob C. Mandell
Department of Radiology, Division of Musculoskeletal Imaging and Intervention, Brigham and Women's Hospital, 75 Francis St, Boston, MA 02115.
Gregory J. Czuczman
Department of Radiology, Division of Musculoskeletal Imaging and Intervention, Brigham and Women's Hospital, 75 Francis St, Boston, MA 02115.
Glenn C. Gaviola
Department of Radiology, Division of Musculoskeletal Imaging and Intervention, Brigham and Women's Hospital, 75 Francis St, Boston, MA 02115.
Varand Ghazikhanian
Department of Radiology, Division of Musculoskeletal Imaging and Intervention, Brigham and Women's Hospital, 75 Francis St, Boston, MA 02115.
Charles H. Cho
Department of Radiology, Division of Neuroradiology, Brigham and Women's Hospital, Boston, MA.

Notes

Address correspondence to J. C. Mandell ([email protected]).

Metrics & Citations

Metrics

Citations

Export Citations

To download the citation to this article, select your reference manager software.

Articles citing this article

Media

Figures

Other

Tables

Share

Share

Copy the content Link

Share on social media