Spinal Interventional Procedures

A. Epidural steroids

Clinical indications and efficacy – An epidural steroid injection (ESI) is typically used to alleviate neck or low back pain that is recalcitrant to more conservative measures. ESI can be especially helpful for patients during an episode of severe pain. Pain relief may also allow or enhance participation in an active rehabilitation program.

Despite the widespread use of ESIs, the medical literature has not established that ESIs are either definitely beneficial or not (Peloso, 2005; Nelemans, 2000). Many early studies and reviews, however, included patients who had had “blind” (non-fluoroscopically-guided) epidurals. The inclusion of such subjects is likely to have biased the literature towards not finding a difference between the experimental (ESI) and control groups, since “blind” epidurals frequently miss their intended targets. More recent reviews limited to ESIs delivered under fluoroscopic guidance, on the other hand, have demonstrated favorable results in groups receiving ESIs (especially for transforaminal lumbar ESIs) in comparison to controls (DePalma, 2005).

In general, patients who are likely to do well have more acute rather than chronic symptoms, and have radicular findings. Response rates are often stated to be ˜80-90% when symptoms have lasted <3 mos, ˜60-80% when <6 mos, and 50% or less at 1 yr or greater, although it should be noted that the rates for spontaneous resolution of pain are also not dissimilarly high. While the optimal timing for ESIs is still unknown, the early use of ESI therapy is generally advocated, rather than delaying therapy.

The interval between injections and the total number of injections allowed within a given time period are subjects of ongoing debate. While the majority of patients respond to steroids in the first few days, some may take up to a week or longer to respond. This can also depend on the type of steroid used (i.e., short-acting vs. long-acting). Moreover, there may be partial suppression of the hypothalamic-pituitary axis for about 2 wks following an ESI. It is thus often advised that repeat injections be considered after at least 2-3 wks have passed.

Generally, repeat injections may be warranted if there is partial relief of symptoms following prior injections. Because addtional relief is not well-documented after a third injection, some clinicians limit the number of ESIs to 3 injections every 6 mos or per year. In practice, the upper limit of ESIs performed in a given year to patients is highly variable, reaching as high as 20 to 40 in some academic and private practice settings according to a recent survey (Cluff, 2002). Patients not responding to the first ESI are sometimes automatically scheduled for 2nd and/or 3rd injections. There is little outcomes-based evidence in the literature supporting this empiric practice. Nonetheless, there may be a rationale to empirically repeating an ESI using a transforaminal approach if the first (ineffective) ESI was administered using an interlaminar or caudal approach.

Mechanism of action – The mechanism of action of ESIs is subject to debate. Proposed mechanisms include: 1) corticosteroid inhibition of phospholipase A2 released by disc injury and 2) a direct action of steroids on the spinal cord, modulating nociceptive input from peripheral nociceptors.

Contraindications and risks – Contraindications to ESIs include: immunocompromised states, infections, increased risk of bleeding (e.g., NSAIDs, clopidogrel, warfarin [INR >1.5], thrombocytopenia [platelets < 50K], coagulopathy), allergy to contrast or injectate, hyperglycemia, adrenal supression, and CHF. NSAIDs should be avoided for 3 days prior to procedures. Plavix and aspirin should be
avoided for 7 days and Coumadin 5 days prior to procedures. Pregnancy is a contraindication to fluoroscopic procedures, although some practitioners may consider “blind” epidurals using the interlaminar approach for severe back pain. Severe central canal stenosis at the level of an injection is also a relative contraindication and the injection should be delivered slowly in this scenario. Interlaminar injections below the level of the stenosis are more desireable.

Risks of ESIs include allergic reaction, complications of steroids (e.g., fluid retention, facial flushing, hyperglycemia, euphoria), infection, epidural bleed, thecal puncture, and spinal cord injury (unlikely). Particulate steroids inadvertently injected into the vasculature have been implicated in central nervous system infarcts (Derby, 2008). Intrathecal steroid administration has been associated with anterior spinal artery syndrome, arachnoiditis, and conus medullaris syndrome, and is best avoided. However, intrathecal steroids have been successfully used to treat intractable post-herpetic neuralgia without report of significant complication (Kotani, 2000).

Ref:

Peloso P, et al. Medicinal and injection therapies for mechanical neck disorders. Cochrane Database Syst Rev 2005;(2):CD000319

;

Nelemans PJ, et al. Injection therapy for subacute and chronic benign LBP. Cochrane Database Syst Rev 2000;(2):CD001824

;

DePalma MJ, et al. A critical appraisal of the evidence for selective nerve root injection in the tx of lumbosacral radiculopathy. Arch Phys Med Rehab 2005;86:1477

;

Cluff R, et al. The technical aspects of epidural steroid injections: a national survey. Anesth Analg 2002;95:403

;

Derby R, et al. Size and aggregation of corticosteroids used for epidural injections. Pain Med 2008;(9):227

;

Kotani N, et al. Intrathecal methylprednisolone for intractable postherpetic neuralgia. NEJM 2000;343:1514.

Technical considerations

The use of fluoroscopy is now strongly recommended to confirm proper needle placement in the epidural space, as it is missed with high frequency in “blind” epidurals, even in experienced hands (25-30% miss rate for lumbosacral ESIs [White, 1980], and higher in cervical ESIs). In addition, the use of contrast media is recommended to confirm placement, and also because negative aspirations for blood can be falsely negative.

The interlaminar ESI (ILESI, also known as translaminar ESI) can take a midline or paramedian path. In the midline approach, the needle traverses through the thick interspinous ligament until it penetrates the ligamentum flavum (often a “pop” is felt) and is in the epidural space. The paramedian approach (˜10-15° off the midline) avoids the interspinous ligament and traverses through the paraspinal muscles and ligamentum flavum. An ILESI is not indicated for patients who has had laminectomy, due to the absence of the ligamentum flavum.

Figure 15a1 Relevant anatomy for the interlaminar approach.

The ILESI is performed at the clinically symptomatic level. In the cervical spine, a typical injectate may be 3-5 ml of 40-60 mg of methylprednisolone mixed in 1-2% lidocaine or saline. In the lumbar spine, a typical injectate may be 6-10 ml of 80-120 mg of methylprednisolone mixed in 1-2% lidocaine or saline.

Autopsy studies have shown that the spinal cord extends caudally no further than L2 for the majority of the population (see figure 15a2). The risk of spinal cord injury due to direct trauma during interlaminar injections is very low, but not non-existent above this level.

The transforaminal ESI (TFESI, aka periradicular injection) delivers the injectate more anteriorly than the ILESI, targeting the nerve root. TFESIs are appropriate for patients with post-laminectomy syndrome. Targeting treatment to right and/or left sides at one or multiple levels is possible. TFESIs are particularly effective for far lateral disc herniations affecting specific nerve roots, since the pathological area can be directly addressed. ILESIs, in contrast, depend on diffusion of the injectate, which may not occur sufficiently enough for clinical relief.

The lumbar TFESI is performed with the C-arm rotated to reveal an oblique view of the spine. Once the “Scotty dog” view is obtained, the C-arm is adjusted until the superior articular process (the ear of the “Scotty dog”) is halfway between the anterior and posterior portion of the vertebral body superior end plate. The superior end plate of the vertebral body should appear superimposed on fluoroscopy. The nerve root passes a few mm inferior to the pedicle and 1-2 mm superficial to the vertebral body. The needle is advanced toward the superior aspect of the neuroforamen, just inferior to the pedicle. With the C-arm rotated to a lateral view, needle is further advanced until the tip is in the dorsal and cephalad quadrant of the neuroforamen, taking care to withdraw the needle slightly if paresthesia is encountered. After a negative aspiration, a small amount of contrast media should be injected to confirm epidural spread and to detect intravascular uptake. Less injectate is used than for ILESIs (about half or less of the interlaminar amounts). Special care, however, should be taken to avoid injection into the artery of Adamkiewicz (which enters the spinal canal in the lower thoracic or lumbar spinal levels near the nerve root), as this can result in a spinal cord infarct.

Figure 15a2 Caudal extent of the spinal cord.

Cervical TFESIs confer the advantage of reducing cephalad spread (sometimes seen with ILESIs, leading to respiratory depression), but are somewhat controversial due to the risk of substantial adverse events (e.g., vascular injection, tetraplegia, hemiplegia), even if they occur only infrequently.

TFESIs are often interchangeably used with selective nerve root blocks (SNRB), although the latter usually refers to injections done to affect a single nerve root without the injectate necessarily reaching the epidural space.

Figure 15a3 TFESI, AP fluoroscopic view (left) and schematic (right).

A caudal ESI can be considered if a lumbar TFESI or ILESI approach is technically difficult. Although caudals are sometimes given for coccydynia, the efficacy for this is unclear. Larger injectate volumes (e.g., 20 ml) are typically used and specific spinal structures not targeted. Thecal puncture risk is lowest with caudals since the thecal sac typically ends at or above S2. Sacral abscess is a known but very unlikely complication.

Local anesthetic is usually given in the area of the sacral cornu, which is easily palpated. A spinal needle with injectate is placed just inferior to the sacral cornu and advanced using fluoroscopic guidance. Contrast dye should be used to observe flow of the injectate to the lower lumbar levels with lateral and A-P views. The injectate will flow in a predominantly cephalad direction toward the lower lumbar levels.

Figure 15a4 Caudal ESI. The needle is in the epidural space.

Ref:

White AJ, et al. Epidural injections for the dx and tx of LBP. Spine 1980;5:78

;

Renfrew DL. Correct placement of epidural steroid injections: fluoroscopic guidance and contrast administration. AJNR 1991;12:1003

;

Rydevik B. Pathoanatomy and pathophysiology of nerve root compression. Spine 1984;9:7.

Figure credits: 15a1, 15a2. Courtesy of Loeser JD, et al., eds. Bonica’s Management of Pain, 3^rd ed. Philadelphia, LWW, 2001, with permission; 15a3. Radiograph courtesy of Ballantyne JC. The MGH Handbook of Pain Management, 3rd ed. Philadelphia, LWW, 2006, with permission. Schematic courtesy of Rathmell JP. Atlas of imaging in regional anesthesia and pain medicine. Philadelphia, LWW, 2005, modified with permission.

B. Zygapophysial (facet) joint injections and RF neurotomy

Anatomy – The zygapophysial joint is also known as the z-joint or facet joint, although use of the latter term is being discouraged. Z-joints are true diarthrodial joints with hyaline cartilage, synovial membranes, and fibrous capsules. The fibrous capsules contain mechanoreceptors and nociceptors, while the subsynovial tissues contain nociceptors. These nociceptors arise from the sympathetic and parasympathetic ganglia.

In the cervical spine, the C2-C3 joint is innervated by two branches of the dorsal ramus of the third cervical nerve: a communicating branch and a medial branch known as the third occipital nerve. Below that in the cervical spine, the innervation of each joint comes from the medial branches of the dorsal rami from each spinal nerve above and below the joint; e.g., the medial branches of C3 and C4 innervate the C3-4 facet. In the lumbar spine, each z-joint is innervated by the medial branch of the dorsal ramus of the spinal nerve at the same level as well as the medial branch from one level rostrally (e.g., the L4-5 z-joint is innervated by medial branches from L3 and L4), with the exception of the L5-S1 z-joint, which in addition to L4 and L5 medial branch nerves may receive additional innervation from S1 medial branch nerve (“triple-innervated”).

Clinical evaluation and efficacy of the interventional therapies – Z-joint pain cannot easily be diagnosed by history, physical exam, or imaging. Historical elements supportive of z-joint pain include pain that worsens with extension (e.g., when standing from a seated position). The exam may be notable for a positive “facet loading maneuver,” but the straight leg raise test is usually negative. Diagnostic blocks of the z-joints or nerves supplying the joints under fluoroscopic
guidance are the only means available to firmly establish the z-joints as pain generators. A small volume injection is necessary to maximize diagnostic integrity. A typical volume for a diagnostic medial branch block is 0.5-1 ml.

If conservative management is unsuccessful, therapeutic injections intraarticularly or near the nervous supply can be considered. Intraarticular steroids and/or local anesthetics can alleviate joint inflammation and wash away potential chemical irritants. A normal joint will accommodate 1.0-1.5 ml of injectate. Some favor only treating the medial branches to avoid potential damage to the joint. Radiofrequency ablation of the innervation is another option. High-grade evidence in the medical literature supporting these interventions, however, is still scant. The quality of the methods and designs of many studies in the literature has been poor, with high variability in the diagnostic criteria and variations in definition of a successful outcome.

Contraindications to z-joint intraarticular injection include INR >1.5, allergies to injectate, active systemic infection or skin lesions, malignancy, recent surgery, and pregnancy (secondary to the need for fluorscopic exposure).

Technique – For cervical z-joint medial branch injections, the patient is placed in the lateral position. For cervical intraarticular injections, it is more difficult to gain joint access from a lateral approach so a posterior approach should be considered. A 25-gauge 1.5 inch needle is used for cervical z-joint and medial branch injections. Care must be exercised to prevent injury to local vascular structures. Use of contrast is adviseable to ensure the needle is in the proper position and to minimize the potential for vascular injection. The C3-7 z-joints are innervated by medial branches that run along the articular pillar in the cervical spine, varying in height and position along the vertebral body. The C2-3 joint is unique as the needle should be placed just lateral to the joint at the location of the large third occipital nerve that innervates the joint. Knowledge of the local anatomy is essential to the performance of these procedures.

Figure 15b1 Z-joint angles in C-Spine, T-spine and L-spine.

Figure 15b2 Lumbar Z-joint injections (fluoroscopic view)

For lumbar z-joint injections, the patient is placed in the prone position. A 22 or 25-gauge 3.5 inch spinal needle is most often used. The ideal entry point for intraarticular injections is at the inferolateral edge of the inferior articulating process, as this is the largest area for injection. From L1-4, the medial branch lies at the intersection of the superior articular process and transverse process. Since the transverse process of S1 is replaced by the sacral ala, L5 ‘medial branch block’ is actually performed as L5 dorsal ramus block, the target being the junction between the sacral ala and the superior articular process.

Radiofrequency (RF) neurotomy/ablation – RF is a high frequency current that generates a well circumscribed spheroidal lesion and thermocoagulates target neural tissue. (Use of the term “rhizotomy” is discouraged, since this refers to cutting a nerve.) RF ablation of the dorsal medial branch (to block all sensory input from the joint) has long been advocated for recalcitrant z-joint pain which has been confirmed with one or more diagnostic blocks. If the z-joint is the primary pain generator for a given patient, symptomatic relief can be expected by denervating the nervous supply to the z-joint. Pain frequently recurs, however, once the medial branch axons regenerate. Larger lesions can reduce recurrence of pain, but also risk excessive local damage, including non-target neural tissue. Local temperature at the electrode tip should be monitored to create safe and predictable lesions. Temperature >90° Celsius will cause tissue to boil and should be avoided. Sensory and motor stimulation prior to RF lesioning is helpful to avoid injury to the motor nerves. Pulsed RF has been proposed as a way to perform RF at lower temperature and without permanent nerve injury.

Overall, the published outcomes for z-joint RF ablation is equivocal. A 2003 Cochrane review revealed limited evidence that RF denervation offers short-term relief for chronic neck pain of z-joint origin and for chronic cervicobrachial pain, and conflicting evidence for its effectiveness for lumbar z-joint pain. Design issues in the research published to date has precluded definitive conclusions. Some recent reviews have questioned the efficacy of radiofrequency neurotomy in treating z-joint mediated pain (Carragee, 2008; Chou, 2009).

Generally, patients carefully selected on the basis of differential dorsal medial branch diagnostic blocks (>80% relief for >1 hr with lidocaine and >2 hrs with bupivacaine) are the best candidates for denervation procedures (Dreyfuss, 2000).

Ref:

Carragee EJ, et al. Treatment of neck pain: injections and surgical interventions: results of the Bone and Joint Decade 2000-2010 Task Force on Neck Pain and Its Associated Disorders. Spine 2008;15:S153.

Chou R, et al.

Only gold members can continue reading. Log In or Register to continue