In 1934 the syndrome of “disc herniation” was born when Mixter and Barr ¹ first proclaimed that a posterior rupture of the intervertebral disc that allowed nuclear material to escape and compress the adjacent spinal nerve root(s) was a common cause of back and leg pain, a condition commonly known as sciatica.

Determining the presence of a disc herniation starts with a careful history. A herniated disc can lead to radiculopathy, radicular pain, or referred pain. Radiculopathy describes neurological changes such as numbness, weakness, and diminished reflexes with or without the presence of pain. Radicular pain is pain that typically follows a dermatomal pattern with or without the presence of sensory or motor deficits. Referred pain is pain that spreads into the lower limbs and is perceived in regions innervated by nerves other than those that innervate the site of noxious stimulation.²

The patient may describe an acute onset of this pain or a period of heavy activity that preceded the development of symptoms, but radiculopathy in the elderly often occurs without a specific inciting event. Sneezing, coughing, or bending usually intensifies the pain. However, in a clinical setting the classic findings of radiculopathy are not always present. Often what emerges is an incomplete pattern of spread or spread that overlaps various dermatomes. Signs and symptoms that could indicate cauda equina syndrome include progressive weakness of the extremity, erectile dysfunction, or bowel or bladder dysfunction.

Anatomy

The lumbar spine serves two basic and vital functions. The first is that it provides the structural framework that houses the spinal cord and spinal nerves. The second is that it provides a mechanical support for the entire axial skeleton. The lumbar spine is composed of five motion segments. Each motion segment consists of the vertebral body, intervertebral disc, pedicles, zygapophyseal joints, posterior lamina, and spinous process. These structural elements are supported by a vast array of ligaments that help maintain the stability of the spine (Fig. 11-1).

Fig. 11-1 Spinal anatomy.

The intervertebral disc has three basic structural components: the nucleus pulposus (NP), annulus fibrosus (AF), and vertebral endplate (VE). The AF and NP are the major structural components of the disc; the VE is responsible for metabolic functions. The NP comprises the gelatinous center of the disc. Under normal conditions the NP equally shares load bearing with the AF. The NP has a high water content, secondary proteoglycans, and aggregate molecules that trap and hold water. Surrounding the NP is the AF, which is fibrous in nature. The annulus is made up of 15 to 25 concentric sheets of collagen known as lamellae. In the normal disc it is the interplay between the nucleus and annulus that allows the disc to distribute very high axial loads equitably.

Nutritional support of the intervertebral disc comes via diffusion of nutrients across the VE. The healthy disc is the largest avascular structure in the body. As a result, it is prone to ischemia in the event of injury to the endplate.

The normal disc is innervated in the outer one third of the annulus by fibers from the sinuvertebral nerve, which also supplies nociceptive input from the dural sac and nerve root sleeve. Branches of the gray rami communicantes have also been identified in the anterolateral annulus.^3,⁴

As the disc degenerates and internal fissures develop, nociceptive fibers may invade into deeper layers of the AF, sometimes reaching the NP itself. These fibers travel along with vascular structures toward the center of the disc.⁵ Once the fissure has penetrated to the nucleus, inflammatory mediators contained within the NP now are exposed to nervous elements. This process, known as internal disc disruption, has been theorized as a major component of chronic low back pain.⁶

A gradual weakening of the outer annulus occurs with aging and endplate disruption. In the normal disc axial loads are distributed evenly across the AF and NP. However, in the case of internal disc disruption, the nucleus no longer evenly shares weight-bearing loads with the annulus. This may cause a generalized disc bulging that is evenly distributed across the annular fibers, or a unilateral bulging. Alone or in combination with other age-related changes (facet arthropathy, ligamentum flavum hypertrophy), this can lead to nerve root compression. Under high mechanical stress the outer annular fibers may disrupt entirely, allowing nuclear material to be displaced outside of the disc itself, resulting in an extruded or sequestered disc herniation (Fig. 11-2).⁷

Fig. 11-2 Pathophysiology of disc-related pain.

Basic Science

The intervertebral disc is immunogenic. This is because, after embryonic development is complete, the avascular NP no longer has exposure to the immune system. Thus, when NP is introduced outside of the confines of the annulus, it is capable of inducing an autoimmune, inflammatory response.⁸

Several inflammatory mediators have been identified in the intervertebral disc in laboratory models designed to duplicate the chemical effects of disc herniation. These inflammatory mediators include phospholipase A₂, prostaglandin E₂, interleukin 1-α, interleukin 1-β, interleukin-6, tumor necrosis factor-α, and nitrous oxide (NO).^9,¹⁰

Animal studies in which autologous NP and AF were placed on or even adjacent to nerve roots demonstrate that it is the nuclear material that induces the chemical radiculitis. The application of NP to nerve roots has been shown to induce axonal and myelin sheets alterations, increased vascular permeability, and decrease intraneural blood flow. Not surprisingly, studies evaluating antagonists to various inflammatory mediators have shown them to abate or prevent hyperalgesic responses in animal models.^11,¹²

There are equally compelling studies involving the effects of mechanical compression. Direct compression of the nerve root both proximal and distal to the dorsal root ganglion (DRG), with or without extruded NP, results in pain-related behaviors in animal models. These pain-related behaviors are affected by a variety of factors, most notably apoptosis.^13,¹⁴ Mechanical compression can produce apoptosis of the more severely injured neurons, which leads to retrograde transport of inflammatory mediators to the DRG, thereby inducing mechanical allodynia. This may be further exacerbated by diminished blood flow caused by edema and direct compression of the neuronal microvasculature. This explains why a disc bulge without extrusion can still produce neuropathic pain.

Imaging

Plain films as a stand-alone tool have a limited role in the diagnosis and management of lumbar radiculopathy, having been supplanted by more advanced imaging. However, in some instances a plain radiograph can help identify bony abnormalities that can lead to radicular pain such as facet arthropathy, spondylolisthesis, and reduced disc height. On a similar note, the role of myelography has changed over the past few decades from the principal means of diagnosing a herniated disc(s) to a secondary role, usually in combination with computed tomography (CT) in patients who cannot undergo magnetic resonance imaging (MRI).

As noted previously, CT is still used today for diagnosing disc herniations, primarily in patients with prior instrumented surgery and those who cannot have MRI (e.g., patients with spinal cord stimulators or pacemakers). In contrast to plain radiographs, CT does allow for imaging of soft tissues and disc contours. The limitations of CT imaging include high radiation exposure and inferior image quality compared to MRI.

MRI remains the modality of choice in diagnosing lumbar disc herniations. Its benefits include excellent resolution of the disc itself and the surrounding nerves, ligaments, and bony structures. The downside is that MRI is relatively sensitive in demonstrating degenerative changes, which can result in the detection of pathology in segments that are not pain generators.¹⁵

The nomenclature for describing disc abnormalities on MRI can be ambiguous. For example, there is no defined set of descriptors for grading or defining lumbar disc herniations. Perhaps more important, when considering whether percutaneous therapies are indicated, there is a need to establish whether or not a disc herniation is contained. MRI evaluations have been shown to be only 70% accurate in making an accurate diagnosis of a contained vs. noncontained disc herniation.¹⁶ Frequently discography can provide valuable information regarding whether or not a disc herniation is contained with the outer fibers of the annulus intact and pathology that is amenable to percutaneous interventions.¹⁷

For the purposes of this chapter, disc herniations are classified as contained and noncontained disc herniations and then divided into the following four categories (Fig. 11-3)⁷:

1 Disc protrusion: The disc protrusion has intact annular fibers and posterior longitudinal ligament (PLL). It is a contained disc herniation.

2 Subannular extrusion: The disc bulge has disrupted the inner annular fibers but has not ruptured the outermost portion of the annulus or the PLL. It is considered a contained disc herniation.

3 Transannular extrusion: The disc material has now extruded past the outer annular fibers and PLL. It is a noncontained disc herniation.

4 Sequestrated disc: This is a free fragment of disc material without a tail into the remainder of the disc. It is a noncontained disc herniation.

Fig. 11-3 Classification of disc pathology.

From Weiner BK, Patel R: The accuracy of MRI in the detection of lumbar disc containment, J Orthop Surg Res 3:46, 2008.

Epidural Steroid Injections (Interlaminar, Transforaminal, and Caudal)

The rationale behind epidural steroid injection (ESI) is that higher concentrations of corticosteroid are delivered to the inflamed nerve root(s) than with oral or parenteral routes, resulting in enhanced pain relief and reduced side effects. ESIs have been studied predominantly in patients with radicular pain (Table 11-1), which is most commonly caused by a disc herniation. There are three ways to access the epidural space: the caudal, interlaminar, and transforaminal approaches. The interlaminar and transforaminal techniques can be used in the cervical, thoracic, and lumbar spine. The caudal epidural space is accessed via the sacral hiatus and thus is reserved for lumbosacral symptomatology.

Table 11-1: Selecting Candidates for Epidural Steroid Injections

Favorable Prognosis	Unfavorable Prognosis
Radiculopathy caused by herniated nucleus pulposus	Degenerative disc disease or spinal stenosis (short duration of benefit)
Short duration of symptoms	Pain >6 months in duration
Leg pain > back pain	Back pain > leg pain
Intermittent pain	Constant pain
Neuropathic descriptors (e.g., shooting, burning, electrical-like)	Nociceptive descriptors (e.g., aching, throbbing, dull)
Absence of psychopathology	Psychological overlay (e.g., somatization disorder, depression, poor coping skills, catastrophization)
Active lifestyle, good physical shape	Sedentary lifestyle, obesity, or deconditioning
Self-employed	On disability or with secondary gain issues
Young age	Older age, concomitant spinal pain generators
Isolated, single-level pathology	Multilevel pathology
No previous or previously successful interventions	Previous spine surgery or multiple failed interventions

Indications/Contraindications

The indications for ESI are the source of ongoing controversy. Historically they have been used to treat conditions as diverse as phantom pain, peripheral neuropathy, complex regional pain syndrome, and pelvic pain. However, their principal use is for the treatment of radiculopathy.

How many ESIs should be performed is also subject to debate. The practice of limiting injections to 3 in 6 months is not based on sound evidence in the form of prospective studies.¹⁸ Rather, the law of diminishing returns and the cumulative risk of steroid-related side effects and complications dictates an inverse relationship between the number of injections and the added benefit for each successive injection. Thus, although limiting the number of injections would appear to be prudent, the exact number and timing of injections need to be determined from clinical trials.

The need for repeat injections should be dictated by the patient’s response to the initial injection. If a patient obtains complete relief with one injection, a second injection can be done on an as-needed basis. If he or she has incomplete relief with the initial injection, a second injection at the same or different site can be repeated in as early as 1 week, although most people wait at least 2 weeks before repeating the procedure. No relief from the first injection warrants reevaluation; if a decision is made to repeat the ESI, either switching the route (i.e., transforaminal instead of interlaminar or intervertebral level) may be beneficial. There are no data to justify an automatic series of ESIs. Most interventionalists limit the use of ESIs to three or four in a 6-month period. The literature suggests that patients who respond to ESI usually require between one and three injections.^18,¹⁹

Risk and Complication Avoidance

The recovery period immediately following ESIs is generally less than 10 minutes. Immediate adverse effects include increased pain from the pressure exerted on the nerves during injection, which typically resolves within hours. Other minor complications include vasovagal reactions, which are more common in females and during cervical spine procedures, weakness or numbness, and transient low back pain. Adverse effects associated with corticosteroids include weight gain, water retention, flushing (hot flashes), mood swings, insomnia, elevated blood sugar levels in patients with diabetes, and suppression of the hypothalamic-pituitary-adrenal axis.

Epidural injections are subject to the same generic complications associated with any procedure such as bleeding, infection, and inadvertent puncture of the adjacent structures (e.g., the dura, kidneys, and blood vessels). Nevertheless, the rare complications that do occur can produce devastating complications such as meningitis, epidural abscess, and epidural hematomas.²⁰

The complication that has received most interest of late is inadvertent spinal cord infarction after transforaminal ESI. There have also been at least 10 cases of this occurring after lumbar transforaminal ESI. Some cases have occurred at levels as low as S1. The most prominent theory as to the etiology of this complication is that there is inadvertent puncture of a spinal radiculomedullary artery. The largest of these vessels, the artery of Adamkiewicz, frequently projects into the lumbar spine as low as the L3 level. Unrecognized injection of particulate steroids into one of these arteries can cause occlusion of blood flow to the spinal cord, leading to infarction. Methods to reduce the risk of this complication include using continuous fluoroscopy or subtraction angiography during injection, minimizing manipulation of the needle, administering a nonparticulate steroid such as dexamethasone, injecting through a catheter inserted through the needle, and using a test dose of local anesthetic before injection.²¹

Epidural hematomas typically develop over a period of hours. The risk of epidural hematoma is less than 2 in 10,000 but increases in the presence of active anticoagulation, bleeding diathesis, and traumatic injection. Epidural hematomas classically present with severe low back pain and spasm along with progressive neurological deterioration. It is critical to recognize this complication within 8 hours because delayed intervention exponentially worsens prognosis.²²

Although rare, there is a small risk of either bacterial meningitis or epidural abscess associated with ESI. The actual infectious risk from ESI is not well known since most literature is in the form of retrospective studies evaluating spinal anesthetics or continuous epidural catheters. The risk of meningitis secondary to intrathecal injection is estimated to be around 1 in 50,000.²²

Anticoagulant Therapy

Regarding anticoagulant therapy, Raj and associates ²³ provided an exhaustive review for the American Society of Interventional Pain Physicians (ASIPP) of not just medication but patient and procedurally-related factors involved in assessing and minimizing the risk of bleeding complications.²³ Horlocker and associates²⁴ were more liberal in arriving at guidelines for the American Society of Regional Anesthesia and Pain Medicine (ASRA). Table 11-2 contains a simplified list of recommendations regarding anticoagulants. These do not address specific patient factors associated with bleeding.

Table 11-2: Anticoagulant Therapy and Neuraxial Procedures

Outcomes Evidence

A key shortcoming in evaluating the evidence for ESI is that many studies evaluating the efficacy of interlaminar injections were done without fluoroscopic guidance. Previous studies have demonstrated high rates (8.8% to 70%) of false loss of resistance for blind (without fluoroscopic guidance) ESIs, which may be higher in the cervical region.^25,²⁶ Even when the epidural space is successfully accessed, blinded injections may not deliver the medication to the area of pathology. In a study conducted in 50 patients with failed back surgery syndrome, Fredman and colleagues²⁶ found that 5 mL of blindly administered injectate reached the targeted area only 26% of the time.

Multiple reviews have been written about the efficacy of ESIs. ²⁷^–³⁰ These reviews are limited by reviewer bias (reviews conducted by people who perform epidural injections tend to be more favorable than those done by people who do not), the inclusion of studies with small sample sizes, serious methodological flaws, inadequate outcome measures, and heterogeneity with respect to route of injection and use of fluoroscopy. In one of the earliest reviews, Koes and associates²⁷ illustrated the difficulty of properly evaluating the literature in a systematic review of 12 randomized clinical trials with disparate methodological qualities, half of which were deemed positive. The primary care physicians who conducted this review concluded that the benefits of epidural steroids, if any, seem to be of short duration. A similar review conducted 4 years later by a French task force of rheumatologists determined that 8 of 13 randomized studies demonstrated no measurable benefit.³¹ The authors of this analysis concluded that no determination could be made regarding the efficacy of epidural steroids for sciatica. The main weaknesses in the studies analyzed in these reviews were that none used fluoroscopic guidance and all used an interlaminar approach, which is probably clinically inferior to transforaminal ESIs.

In recent European guidelines for the management of low back pain, Airaksinen and colleagues ³² concluded that epidural corticosteroid injections should be considered only for radicular pain, if a contained disc prolapse is the cause of the pain, and if the corticosteroid is injected close to the site of pathology. They further noted that injections should be fluoroscopically guided toward the ventral epidural space. These recommendations are in direct contrast with those outlined in a recent report by a subcommittee of the American Academy of Neurology that concluded that lumbosacral ESIs for radicular pain do not improve function, provide long-term pain relief (>3 months), or obviate the need for surgery.³³ The authors found insufficient data to draw a conclusion for cervical ESI.

Recent systematic reviews by Abdi and associates ²⁸ and Boswell and associates³⁴ reached different conclusions. The authors found strong evidence for short-term pain relief and functional improvement for lumbar transforaminal, cervical interlaminar, and caudal ESI and moderate evidence for long-term relief. The evidence supporting lumbar interlaminar injections was strong for short-term improvement but limited for long-term benefit.

The randomized controlled trials that have evaluated caudal, interlaminar, and transforaminal ESIs (TFESIs) to date carry major flaws in that fluoroscopic guidance was not used in most of the early studies and most were conducted in patients hospitalized for their pain.

One small randomized controlled study conducted by an orthopedic group found that lumbar TFESIs decreased the need for decompression surgery for up to 5 years since most of the patients who underwent lumbar TFESI elected not to undergo decompression surgery.³⁵

Fluoroscopic guidance and method of injection (interlaminar vs. transforaminal) appear to account for most of the heterogeneity among the systematic reviews. The reviews with heterogeneity among methods of injection did not find a clinical benefit for the procedure,^26,³³ whereas those with stratified trials based on the method of injection found evidence to support transforaminal and caudal ESIs performed with fluoroscopic guidance.^28,³⁴ Considering that transforaminal and caudal ESI more reliably deliver medication to the ventral epidural space, this finding is not surprising.²⁵ Transforaminal injections have also been found to be clinically superior to interlaminar ESI in two head-to-head comparisons.^36,³⁷ Transforaminal ESIs appear to afford better and longer-lasting relief than interlaminar and caudal ESI; however, the added benefit must be balanced against the higher risk.